comfytts-models / MANIFEST.json
Prompt-Pirate's picture
Upload folder using huggingface_hub
66b5948 verified
Raw History Blame Contribute Delete
151 kB
{
"repo": "Prompt-Pirate/comfytts-models",
"generated": "2026-08-29T13:14:06.223Z",
"engines": {
"irodori": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/irodori",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (Irodori-TTS-v4-Small-GGUF; original Aratako/Irodori-TTS-v4.1-Small)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/Irodori-TTS-v4-Small-GGUF"
},
"licence": {
"weights": "MIT",
"verified": "HF model card Aratako/Irodori-TTS-v4.1-Small license:mit + audio-cpp/audio.cpp-gguf card row 'MIT' (2026-08-23); codec Aratako/Semantic-DACVAE-Japanese-32dim MIT; model card adds non-binding ethical-use notes (no impersonation without consent)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 1 (sha256 4c30f25aaeb7c273194ee11e25c2ca3e7abde4c09dba8d86b1727528a57bfdd8, 1,762,148,352 bytes)"
],
"notes": "LOCAL USE via the audio.cpp route (v25.82.0, wave 1 of the exclusives lane): irodori-tts-cpp / irodori-clone-cpp / irodori-design-cpp all on ONE f16 v4-Small GGUF from audio-cpp's own HF repo. NOT mirrored by us (owner: 'if it's available do it' - local use; the licence law governs redistribution only).",
"files": []
},
"miotts": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/miotts",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (MioTTS-1.7B-GGUF + MioCodec-25Hz-44.1kHz-v2-GGUF; original Aratako/MioTTS-1.7B + Aratako/MioCodec-25Hz-44.1kHz-v2)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/MioTTS-1.7B-GGUF"
},
"licence": {
"weights": "Apache-2.0 (MioTTS-1.7B) + MIT (MioCodec)",
"verified": "HF model cards Aratako/MioTTS-1.7B license:apache-2.0 and Aratako/MioCodec-25Hz-44.1kHz-v2 license:mit + the audio-cpp card rows (2026-08-23)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 1 (miotts-1.7b-orig.gguf sha256 29f5dcfbf5a94e6ae6e2574c59cd6730e3b4c868436d5d455e4d943e121e46ed, 3,518,532,512 bytes; miocodec-25hz-44khz-v2-orig.gguf sha256 966ff5516355f1c1677e2443bb94bb81ec6608590747883574fe8958e2d6dabc, 905,706,656 bytes)",
"the codec is a COMPANION model passed as --session-option miotts.codec_model_path (manifest audiocpp.companions)"
],
"notes": "LOCAL USE via the audio.cpp route (v25.82.0, wave 1): miotts-clone-cpp, a pure cloner (--voice-ref required, no presets, no transcript), en/ja, orig precision (the only 'pass' file). NOT mirrored by us. WAVE E (v25.104.0, 2026-08-26): miocodec-vc-cpp - an STS voice changer whose MAIN model is the MioCodec file wave 1 fetched as the miotts COMPANION (MIT per the audio-cpp card row for MioCodec-25Hz-44.1kHz-v2-GGUF). Non-deterministic by measurement (no seed declared); the family's s2s task lands in wave F. No new files. WAVE F (v25.105.0, 2026-08-26): miocodec-s2s-cpp - the s2s sibling of the wave-E voice changer on the same wave-1 file (two conversion algorithms on one codec, A/B by ear). Non-deterministic like its sibling. No new files.",
"files": []
},
"confucius4": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/confucius4",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (Confucius4-TTS-GGUF; original netease-youdao/Confucius4-TTS)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/Confucius4-TTS-GGUF"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF model card netease-youdao/Confucius4-TTS license:apache-2.0 + audio-cpp/audio.cpp-gguf card row 'Apache-2.0' (2026-08-23)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 2 (confucius4-tts-orig.gguf sha256 eec4ab3fae3cda1e8ec2b96599442def30dbc980ea817f42b94e073158536244, 8,192,757,760 bytes)"
],
"notes": "LOCAL USE via the audio.cpp route (v25.83.0, wave 2 of the exclusives lane): confucius4-clone-cpp, a 14-language TRANSCRIPT-FREE cloner (--voice-ref required; --reference-text ignored); en/zh are upstream's validated text frontends. UPSTREAM STATUS: EXPERIMENTAL (the binary's own model_specs). Original-dtype GGUF (the only packaged precision). NOT mirrored by us.",
"files": []
},
"mosslocal": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/mosslocal",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (MOSS-TTS-Local-v1.5-GGUF; original OpenMOSS-Team/MOSS-TTS-Local-Transformer-v1.5 + the MOSS audio tokenizer, bundled in ONE gguf)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/MOSS-TTS-Local-v1.5-GGUF"
},
"licence": {
"weights": "Apache-2.0",
"verified": "audio-cpp/audio.cpp-gguf card row 'Apache-2.0' for MOSS-TTS-Local-v1.5-GGUF + the originals list naming OpenMOSS-Team/MOSS-TTS-Local-Transformer-v1.5 (2026-08-23)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 2 (moss-tts-local-v1.5-bf16.gguf sha256 9ff6c86f3d4cfbd227cc1b7618e3e016aaa129686f9939fe422a49e0e7369415, 13,367,797,280 bytes; bf16 chosen over the binary's recommended q8_0 because upstream docs/gguf.md marks 16-bit a clean Pass and q8_0 'Pass (ASR match, drift)')"
],
"notes": "LOCAL USE via the audio.cpp route (v25.83.0, wave 2): mosslocal-clone-cpp (31 languages + code-switching; an OPTIONAL reference transcript rides a sidecar or is omitted) and, if the probe shipped it, mosslocal-tts-cpp (text-only, model-chosen voice). A DIFFERENT model family from the ComfyUI `moss` (v1 8B delay) and `mossnano` kinds. NOT mirrored by us.",
"files": []
},
"glmtts": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/glmtts",
"junction": null,
"gate": null,
"origin": {
"repo": "mirek190/audio.cpp (Text to audio (TTS)/GLM-TTS_Q8.gguf - the COMMUNITY mirror the binary's own model_specs name; original zai-org/GLM-TTS)",
"url": "https://huggingface.co/mirek190/audio.cpp/tree/main/Text%20to%20audio%20(TTS)"
},
"licence": {
"weights": "MIT",
"verified": "HF model card zai-org/GLM-TTS license:mit (2026-08-23); the mirror repo's own card says apache-2.0 - that is the MIRROR's claim, the original card governs; the GGUF's provenance is community (pinned by sha)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 3 (GLM-TTS_Q8.gguf sha256 28e2b66333fd5f051466520f6a8a5f26cba7f3b768891006168a4560972107cb, 5,143,813,728 bytes)"
],
"notes": "LOCAL USE via the audio.cpp route (v25.84.0, wave 3 of the exclusives lane): glmtts-clone-cpp, a zh/en zero-shot cloner that REQUIRES the reference's exact transcript (the first consumer of the sidecar-or-omitted transcript rule). Mixed Q8_0/F16 GGUF, upstream status community. NOT mirrored by us.",
"files": []
},
"outetts": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/outetts",
"junction": null,
"gate": null,
"origin": {
"repo": "mirek190/audio.cpp (Text to audio (TTS)/Llama-OuteTTS-1.0-1B_Q8.gguf - the COMMUNITY mirror the binary's own model_specs name; original OuteAI/Llama-OuteTTS-1.0-1B)",
"url": "https://huggingface.co/mirek190/audio.cpp/tree/main/Text%20to%20audio%20(TTS)"
},
"licence": {
"weights": "CC-BY-NC-SA-4.0 + Llama 3.2 Community License - NON-COMMERCIAL",
"verified": "HF model cards OuteAI/Llama-OuteTTS-1.0-1B and OuteAI/Llama-OuteTTS-1.0-1B-GGUF license:cc-by-nc-sa-4.0 (2026-08-23); the base model is Llama 3.2 (its Community License applies too); the mirror repo's own card says apache-2.0 - that is the MIRROR's claim, the original cards govern"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 3 (Llama-OuteTTS-1.0-1B_Q8.gguf, 3,029,895,456 bytes; the Qwen3 Forced Aligner + DAC codec are embedded in the same file)"
],
"notes": "LOCAL USE ONLY via the audio.cpp route (v25.84.0, wave 3): outetts-clone-cpp (23 languages; reference ~10 s, max 20 s, transcript REQUIRED for cloning) and, if the probe shipped it, outetts-tts-cpp (text-only, model-chosen voice). NON-COMMERCIAL weights, built on the owner's ruling that availability decides local use; never mirrored, never redistributed by us.",
"files": []
},
"vevo2": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/vevo2",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (Vevo2-GGUF; original RMSnow/Vevo2)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/Vevo2-GGUF"
},
"licence": {
"weights": "CC-BY-NC-ND-4.0 - NON-COMMERCIAL, NO DERIVATIVES",
"verified": "HF model card RMSnow/Vevo2 license:cc-by-nc-nd-4.0 + the audio-cpp card row (2026-08-23)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 4 (vevo2-orig.gguf sha256 8f4d678b08faba711251905c032379a9f763fff942454739f407b3dc07ddc606, 7,449,615,488 bytes; orig over the binary's default q8_0 because upstream docs/gguf.md marks q8_0 'No (mixed route drift)')"
],
"notes": "LOCAL USE ONLY via the audio.cpp route (v25.85.0, wave 4): vevo2-clone-cpp = the zero-shot TTS route (reference + OPTIONAL exact transcript; upstream warns a mismatched transcript can leak reference content). v25.88.0 (the STS lane): vevo2-vc-cpp = the voice-changer route (style_preserved_vc) on the SAME file - a second workflow on this kind. svc / s2s / singing remain future lanes. NON-COMMERCIAL + NO-DERIVATIVES weights; never mirrored, never redistributed by us. WAVE F (v25.105.0, 2026-08-26): vevo2-svc-cpp - singing style-preserved conversion on the SAME wave-4 orig bundle (the family's third route). The family's s2s task (speech EDITING into target text) is validator-legal since v25.105.0 but deliberately NOT shipped - it needs a text row on the STS surfaces. Same CC-BY-NC-ND weights, local only, no new files.",
"files": []
},
"inflect": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/inflect",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (Inflect-Micro-v2-GGUF; original owensong/Inflect-Micro-v2)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/Inflect-Micro-v2-GGUF"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF model card owensong/Inflect-Micro-v2 license:apache-2.0 + the audio-cpp card row (2026-08-23). RUNTIME DEPENDENCY: eSpeak-ng (GPL-3.0-or-later) - the engine phonemises through an external espeak-ng.dll + espeak-ng-data that the app places beside its models, copied from the NeuTTS 1.4.1 wheel already vendored in parrot-tts-nodes; a separate process dependency of the audio.cpp CLI, never linked into Parrot"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 5 (inflect-micro-v2-orig.gguf sha256 d4af1cb6a92cdd8be550e8e7c25805ece222ec0f8e75daf26fc00b4e04ef4b03, 72,082,176 bytes) + the espeak-ng companions COPIED from the vendored NeuTTS wheel (espeak-ng.dll sha256 02d7d1182b7c753c0a08f65e5c180af21e07ba3a3a948f44c24040fea69590ae, 397,824 bytes; espeak-ng-data 371 files, marker phontab) into models/audiocpp/espeak-ng/"
],
"notes": "LOCAL USE via the audio.cpp route (v25.87.0, the rider): inflect-tts-cpp, ONE synthetic English voice (9.4M VITS, 72 MB). Apache-2.0 weights, community GGUF from audio-cpp's own repo; NOT mirrored (the espeak-ng GPL dependency would ride along with any redistribution - a deliberate exclusion).",
"files": []
},
"seedvc": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/seedvc",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (SeedVC-MLX-GGUF; original Plachta/Seed-VC via mlx-community/SeedVC-MLX)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/SeedVC-MLX-GGUF"
},
"licence": {
"weights": "GPL-3.0",
"verified": "HF model cards Plachta/Seed-VC and mlx-community/SeedVC-MLX license:gpl-3.0 + the audio-cpp card row (2026-08-23)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 6 (seed-vc-mlx-f16.gguf sha256 03740be5b4b55ae677c34d63514ff879aaedd6a77fd31773938166fb84debf93, 3,629,186,560 bytes)"
],
"notes": "LOCAL USE via the audio.cpp route (v25.88.0, the STS lane part 1): seedvc-vc-cpp, speech-to-speech voice conversion (the V2 route) - a source clip + the character's bound reference as the target. GPL-3.0 weights: local use by the owner's ruling, NEVER mirrored or redistributed by us. WAVE F (v25.105.0, 2026-08-26): seedvc-svc-cpp - the registry's first SINGING voice conversion route, on the SAME wave-6 f16 file. f0_condition is pinned true at the block level: without it every svc render crashes the 0.6.1 CUDA build (GGML_ASSERT, probed 12/12). Same GPL-3.0 weights, no new files.",
"files": []
},
"bsroformer": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/bsroformer",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (BS-RoFormer-ep368-GGUF; original ZFTurbo/Music-Source-Separation-Training model_bs_roformer_ep_368_sdr_12.9628)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/BS-RoFormer-ep368-GGUF"
},
"licence": {
"weights": "Apache-2.0",
"verified": "the audio-cpp/audio.cpp-gguf card row for BS-RoFormer-ep368-GGUF (2026-08-24)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 8 (bs-roformer-ep368-q8_0.gguf sha256 9a55a8cad369d00f6e0fb208bb0cd87e30e25430772b8491e20a4eace6423ad2, 172,532,256 bytes)"
],
"notes": "SEPARATION MODEL, NOT A RENDER ROUTE (v25.90.0, the Separation lane): bs-roformer-cpp in the separation registry root audiocpp-adapter/sep/ - 2 stems (vocals + instrumental), the Clean Vocals / Refiner 🧹 audio.cpp choice's 'pass' model. Apache-2.0 - kept source origin like the ASR rows (audio-cpp's own repo is the ungated home).",
"files": []
},
"melroformer": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/melroformer",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (Mel-Band-RoFormer-GGUF; the KimberleyJensen Mel-Band RoFormer family)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/Mel-Band-RoFormer-GGUF"
},
"licence": {
"weights": "MIT",
"verified": "the audio-cpp/audio.cpp-gguf card row for Mel-Band-RoFormer-GGUF (2026-08-24)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 8 (mel-band-roformer-f16.gguf sha256 cbdf174dc716615dcd304678f67fc1d2ac6286fbc647e72837b5a6d447a07add, 456,514,816 bytes)"
],
"notes": "SEPARATION MODEL, NOT A RENDER ROUTE (v25.90.0): mel-band-roformer-cpp - 2 stems; audio.cpp flags this family's GGUFs as 'drift' (f16 chosen as the closer one); the owner compares it with the venv Mel-Band by ear. MIT - kept source origin.",
"files": []
},
"htdemucs": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/htdemucs",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (HTDemucs-GGUF; original facebookresearch/demucs)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/HTDemucs-GGUF"
},
"licence": {
"weights": "MIT",
"verified": "the audio-cpp/audio.cpp-gguf card row for HTDemucs-GGUF (2026-08-24); facebookresearch/demucs is MIT"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 8 (htdemucs-f16.gguf sha256 f8c54ba35df95aafea7881eac214b58ec18a73cbcb6ff5a90f4b762631a286b6, 84,059,168 bytes)"
],
"notes": "SEPARATION MODEL, NOT A RENDER ROUTE (v25.90.0): htdemucs-cpp - 4 stems (drums/bass/other/vocals), the 🥁 Split Stems tab's model; needs 44.1 kHz input (the app decodes). MIT - kept source origin.",
"files": []
},
"rvc": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/rvc",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (RVC-GGUF; Retrieval-based Voice Conversion - the base HuBERT + RMVPE + packaged v2 default voice)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/RVC-GGUF"
},
"licence": {
"weights": "MIT",
"verified": "the audio-cpp/audio.cpp-gguf card row for RVC-GGUF (2026-08-24); RVC upstream is MIT"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 9 (rvc-f16.gguf sha256 5ddfc32ab63199bcca4bcd95a007a8add35ccff7e2c20e1272f19b9617410b57, 1,260,188,416 bytes)"
],
"notes": "LOCAL USE via the audio.cpp route (v25.91.0, STS part 2): rvc-vc-cpp - voice conversion into a VOICE MODEL (the owner's own .pth/.index files under models\\voices\\rvc\\<Voice>\\ or a character's _refined\\.rvc-model\\ - never part of any mirror), server mode. The kind `rvc` existed as the Applio/RVC TRAINING lane's kind; this is the first engine that RUNS it. MIT base - kept source origin (audio-cpp's own repo is the ungated home; user voices are the owner's files).",
"files": []
},
"qwen3asr": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/qwen3asr",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (Qwen3-ASR-1.7B-GGUF + Qwen3-ForcedAligner-0.6B-GGUF; original Qwen/Qwen3-ASR-1.7B, Qwen/Qwen3-ForcedAligner-0.6B)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/Qwen3-ASR-1.7B-GGUF"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF model cards Qwen/Qwen3-ASR-1.7B and Qwen/Qwen3-ForcedAligner-0.6B license:apache-2.0 (2026-08-23); the Silero VAD companion file is MIT (snakers4/silero-vad, redistributed inside the audio.cpp repo)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 7 (qwen3-asr-1.7b-q8_0.gguf sha256 da4fc2ac7f24dee784d1684eb1f35836cdbf559519452ae11777670734c0a4f8, 2,473,010,048 bytes; the aligner companion qwen3-forced-aligner-0.6b-q8_0.gguf sha256 75209490b11cec2b0db749ca5f4ff92266f58efd30f7fd04d9eb2a3ac9cc929f, 1,129,966,496 bytes; silero_vad_16k.safetensors sha256 c59271c284ae9c8335d795d60e0bfdb71aaaceec578d9bd9ffc1b8153c319ea1, 1,239,748 bytes)"
],
"notes": "ASR MODEL, NOT A RENDER ROUTE (v25.89.0, the ASR lane): qwen3-asr-cpp in the ASR registry root audiocpp-adapter/asr/ - the Transcribe tab's + the Refiner sidecars' audio.cpp backend, 30 languages incl. ja/zh/ko. Word timestamps need the aligner + VAD companions (the 0.6.1 binary refuses --words-out without them). Apache-2.0 weights - mirror-eligible in principle; kept source origin because audio-cpp's own repo is the ungated home and nothing in this row is needed by the render installer.",
"files": []
},
"parakeet": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/parakeet",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (Parakeet-TDT-0.6B-v3-GGUF; original nvidia/parakeet-tdt-0.6b-v3)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/Parakeet-TDT-0.6B-v3-GGUF"
},
"licence": {
"weights": "CC-BY-4.0",
"verified": "HF model card nvidia/parakeet-tdt-0.6b-v3 license:cc-by-4.0 (2026-08-23)"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 7 (parakeet-tdt-0.6b-v3-q8_0.gguf sha256 074e61ac1abd3d3efcfc10798d17bf9a975b31768b466fc2578364d206dde64c, 915,733,744 bytes)"
],
"notes": "ASR MODEL, NOT A RENDER ROUTE (v25.89.0, the ASR lane): parakeet-tdt-cpp in the ASR registry root - the CPU transcription model (11x realtime on the CPU build), 25 European languages, native word timestamps. CC-BY-4.0 (attribution) - mirror-eligible in principle; kept source origin like qwen3asr.",
"files": []
},
"zonos2": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/zonos2",
"repoPath": "engines/zonos2",
"junction": {
"link": "ComfyUI\\models\\zonos2"
},
"gate": null,
"origin": {
"repo": "Zyphra/Zonos-v0.1-hybrid + Zyphra/Zonos-v0.1-speaker-embedding",
"url": "https://huggingface.co/Zyphra/Zonos-v0.1-hybrid"
},
"licence": {
"weights": "Apache-2.0",
"verified": "cardData both repos; zonos2-clone-v1 manifest README section Licence"
},
"postInstall": [
"pack: Saganaki22/Zonos2_TTS-ComfyUI v0.1.7 (standalone pack, not tts_audio_suite)",
"the zonos2 loader has NO searching branch — the ComfyUI-side folder MUST be a junction to this dir"
],
"notes": "localPath resolved and verified non-empty 2026-08-18 (dac_44khz/, speaker_encoder/, two safetensors checkpoints).",
"files": [
{
"path": "dac_44khz/config.json",
"sha256": "4eb55fb9af1990b8d608184ad29b70e358589719af7ea8d3c06998f7c2264a64",
"bytes": 541,
"url": null
},
{
"path": "dac_44khz/model.safetensors",
"sha256": "6128ebff483a41422b0164d079a3773b0d8d82e64c4293d775994cbf8baf913a",
"bytes": 306507276,
"url": null
},
{
"path": "dac_44khz/preprocessor_config.json",
"sha256": "c7d295758ce5777d6d88fef1996e94adc8ef3e2237ddfc5ecc24d1407aaddd7d",
"bytes": 206,
"url": null
},
{
"path": "speaker_encoder/config.json",
"sha256": "297ac64afea59191e5aa446cb8acfdfdecfcd34771edda47d6948d1be5834ae9",
"bytes": 968,
"url": null
},
{
"path": "speaker_encoder/configuration_ecapa_tdnn.py",
"sha256": "6e187fd0adb8245829c855614e880551e2ec14c2372b2e5ad3c7e6565726d860",
"bytes": 2917,
"url": null
},
{
"path": "speaker_encoder/feature_extraction_ecapa_tdnn.py",
"sha256": "4a353b0dc0171f640dc367f2e55e1ede045195d5d2134fa6124642ff3eb74927",
"bytes": 4836,
"url": null
},
{
"path": "speaker_encoder/model.safetensors",
"sha256": "df60a638e7f4a29331c0af2bd2984ee5b992fee9d5923c776f7e4bdc3dedea48",
"bytes": 24010000,
"url": null
},
{
"path": "speaker_encoder/modeling_ecapa_tdnn.py",
"sha256": "88281ea40ad4792943d598d416476faa4b311acedf153350a556ceaa6b55805a",
"bytes": 10343,
"url": null
},
{
"path": "speaker_encoder/preprocessor_config.json",
"sha256": "c67e710a6d1609f774e112459e7b47638ca185ae1b147d2e790c45c9cde3652f",
"bytes": 275,
"url": null
},
{
"path": "speaker_encoder/tokenizer_config.json",
"sha256": "30ac911d59755bbaa7f0986d8cbd4cc98ff73f4752bae8829936a573e663d6b0",
"bytes": 165,
"url": null
},
{
"path": "speaker_encoder/tokenizer_ecapa_tdnn.py",
"sha256": "299cf9528498ef093e6cfbb88d8738368c133342362e2cbe52e99ef59e39b13e",
"bytes": 2627,
"url": null
},
{
"path": "zonos2-bf16.safetensors",
"sha256": "cd2fd4c9e867750e27ccd0544d2ff4cc86b5c403d6bff4ceb7540068d31b26b4",
"bytes": 15336294640,
"url": null
},
{
"path": "zonos2-fp8-mixed.safetensors",
"sha256": "326e1d68c66cde0af14c27570edfdbf2d339be50ce3d6fafb8778ceaf3f9381a",
"bytes": 10504655360,
"url": null
}
]
},
"voxcpm": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/VoxCPM",
"repoPath": "engines/voxcpm",
"junction": {
"link": "ComfyUI\\models\\tts\\VoxCPM"
},
"gate": null,
"origin": {
"repo": "OpenBMB/VoxCPM2",
"url": "https://huggingface.co/openbmb/VoxCPM2"
},
"licence": {
"weights": "Apache-2.0",
"verified": "voxcpm2-clone-v1 manifest notes: 'Weights Apache-2.0, ungated'"
},
"postInstall": [
"pack: Saganaki22's VoxCPM2 fork v0.3.1 (standalone pack)",
"the actual weights sit one level down at VoxCPM\\VoxCPM2\\ — the parent dir is the folder-key root, only that one subfolder exists inside it"
],
"notes": "localPath resolved and verified non-empty 2026-08-18 (contains exactly one subfolder, VoxCPM2\\, with model.safetensors + audiovae.pth + tokenizer files). Junction link is lowercase 'tts' DELIBERATELY — the pack registers the case-mismatched folder key extra_model_paths.yaml can never reach (CLAUDE.md, voxcpm2 roster row). WAVE C (v25.102.0, 2026-08-26): audio.cpp GGUF twins exist (voxcpm2-clone-cpp + the voxcpm2-tts-cpp text-only roller; VoxCPM2 ORIG GGUF from audio-cpp/audio.cpp-gguf, sha-locked wave 11). Same Apache-2.0 weights in a GGUF container; NOT staged in OUR mirror - the audio-cpp repo is already their ungated home.",
"files": [
{
"path": "VoxCPM2/README.md",
"sha256": "7384fad93ce2d98f47d5c3170597f3b31d414c12c92e7fdf3121fa90f19fe29d",
"bytes": 7776,
"url": null
},
{
"path": "VoxCPM2/audiovae.pth",
"sha256": "94b5d51e107e0507d4acc976cfdadb64edd6fd06d1f751dadbf2fd1594274bf1",
"bytes": 376951122,
"url": null
},
{
"path": "VoxCPM2/config.json",
"sha256": "405f0dcd92f7feba6011ed4eac5c8d4f74cba9712f07fd5cfa3063bbdd95402c",
"bytes": 4336,
"url": null
},
{
"path": "VoxCPM2/model.safetensors",
"sha256": "f7f964cfa9da23653baec6e6f7750719977ad944ed9f95fe52fe3a620506891d",
"bytes": 4580080592,
"url": null
},
{
"path": "VoxCPM2/special_tokens_map.json",
"sha256": "068594063e37662c02b21acf42ebb334ef6a74fb810e68a2368f88f08351de76",
"bytes": 1632,
"url": null
},
{
"path": "VoxCPM2/tokenization_voxcpm2.py",
"sha256": "84489ea32b6ee0cae22ed5480cacb6df85c46624c3119be9a2021c3649a12729",
"bytes": 2895,
"url": null
},
{
"path": "VoxCPM2/tokenizer.json",
"sha256": "f8984687e4a92a3503d521396d454b7d68e9fdaab2a0288eb3536c7c1aa4bc20",
"bytes": 3676772,
"url": null
},
{
"path": "VoxCPM2/tokenizer_config.json",
"sha256": "e78a3ebb48a0b9437efd1823b6b726c823da89e49dd8bcc90c02419d9baa772b",
"bytes": 5059,
"url": null
}
]
},
"firered2": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/FireRedTTS2",
"repoPath": "engines/firered2",
"junction": {
"link": "ComfyUI\\models\\TTS\\FireRedTTS2"
},
"gate": null,
"origin": {
"repo": "FireRedTeam/FireRedTTS2",
"url": "https://huggingface.co/FireRedTeam/FireRedTTS2"
},
"licence": {
"weights": "Apache-2.0",
"verified": "FireRedTeam/FireRedTTS2 cardData (firered2-clone-v1 manifest notes)"
},
"postInstall": [
"pack: 1038lab/ComfyUI-FireRedTTS (GPL-3.0 pack; weights Apache-2.0)",
"models are PACK-GLOBAL — only parrot-hooks' ParrotEngineUnload or a full ComfyUI restart frees them, never ⏏ /free alone"
],
"notes": "localPath resolved and verified non-empty 2026-08-18 (codec.pt, config_codec.json, config_llm.json, llm_posttrain.pt, llm_pretrain.pt, Qwen2.5-1.5B\\, README.md). Junction needed because the pack hardcodes models_dir/TTS/FireRedTTS2 with no searching branch.",
"files": [
{
"path": "Qwen2.5-1.5B/LICENSE",
"sha256": "832dd9e00a68dd83b3c3fb9f5588dad7dcf337a0db50f7d9483f310cd292e92e",
"bytes": 11343,
"url": null
},
{
"path": "Qwen2.5-1.5B/README.md",
"sha256": "84cc3bdbfd781149913126b70981cf1e341176fe4bc9d878601a83f18b035ddf",
"bytes": 3850,
"url": null
},
{
"path": "Qwen2.5-1.5B/config.json",
"sha256": "0e8c8aa86468aba09c9d32157ff4bc2301c7e6c50e4398960425b2ea71e66f77",
"bytes": 684,
"url": null
},
{
"path": "Qwen2.5-1.5B/generation_config.json",
"sha256": "8c970692323e3ea0e9b8b0a4dca79388d31226e41f83c9fd6014804280ebf6e8",
"bytes": 138,
"url": null
},
{
"path": "Qwen2.5-1.5B/merges.txt",
"sha256": "599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3",
"bytes": 1671839,
"url": null
},
{
"path": "Qwen2.5-1.5B/original_repo_id.json",
"sha256": "073a8264cf882196e9478c0d0019a00a29a6315bbb826bd6a85dfd38ce1ef3f3",
"bytes": 38,
"url": null
},
{
"path": "Qwen2.5-1.5B/tokenizer.json",
"sha256": "c0382117ea329cdf097041132f6d735924b697924d6f6fc3945713e96ce87539",
"bytes": 7031645,
"url": null
},
{
"path": "Qwen2.5-1.5B/tokenizer_config.json",
"sha256": "c91efca15ceff6e9ee9424db58a6f59cd41294e550a86cbd07e3c1fb500b34f9",
"bytes": 7228,
"url": null
},
{
"path": "Qwen2.5-1.5B/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "README.md",
"sha256": "80e52ddc3813bc7bea3252654be408dca1a439a26afdd4f82aa04b8b4f6ffa67",
"bytes": 8883,
"url": null
},
{
"path": "_fetch-results.txt",
"sha256": "cde8544d636d5efe8776188486e3128d328f26c07435f1c2b6d1a0aa0875b270",
"bytes": 1400,
"url": null
},
{
"path": "codec.pt",
"sha256": "5e0a31e0390d69e026f48e908eb5e74393c71e10f773998fcdac61abd60f5e21",
"bytes": 4298863130,
"url": null
},
{
"path": "config_codec.json",
"sha256": "4f3ca8a620ca00f7fb46bda95fbd2d5640e2ddbd2d14be71eac10d879b38b5d9",
"bytes": 1593,
"url": null
},
{
"path": "config_llm.json",
"sha256": "3acf4395157d694f994df1583c9fe012e6e0a27bee30a0025b64d515bfdab140",
"bytes": 277,
"url": null
},
{
"path": "llm_posttrain.pt",
"sha256": "70f70632c028bb0396d479a340115d01e5f7a952451f283319a549d285ea0a7f",
"bytes": 8269965409,
"url": null
},
{
"path": "llm_pretrain.pt",
"sha256": "249681daf85dabb1d580913b27a9a4a82aa6f79d2cdfda0f955578aa70a8efd4",
"bytes": 8269965345,
"url": null
}
]
},
"longcat": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/audiodit",
"repoPath": "engines/longcat",
"junction": {
"link": "ComfyUI\\models\\audiodit"
},
"gate": null,
"origin": {
"repo": "meituan-longcat/LongCat-Audio-Codec (AudioDiT) + google/umt5-base",
"url": "https://huggingface.co/meituan-longcat"
},
"licence": {
"weights": "MIT",
"verified": "longcat-clone-v1 manifest: 'Pack and weights MIT'"
},
"postInstall": [
"pack: ComfyUI-LongCat-AudioDIT-TTS (was already installed on this machine, unmanifested until discovered)",
"the loader has NO searching branch — models\\audiodit junction is mandatory"
],
"notes": "localPath is the PARENT dir on purpose — it holds two siblings the engine needs together: LongCat-AudioDiT-3.5B-bf16\\ (the DiT checkpoint) and umt5-base-tokenizer\\ (the text encoder). Verified non-empty 2026-08-18.",
"files": [
{
"path": "LongCat-AudioDiT-3.5B-bf16/README.md",
"sha256": "23714d6e3d441b596fb93046bed187c005b7e9bee1115c05874766ede2db7c72",
"bytes": 4365,
"url": null
},
{
"path": "LongCat-AudioDiT-3.5B-bf16/config.json",
"sha256": "1b7e28a41b26758b74b31fda281bd98fbec5be25a6ddd9acadce086c32603580",
"bytes": 2469,
"url": null
},
{
"path": "LongCat-AudioDiT-3.5B-bf16/model.safetensors",
"sha256": "dd3beb8dd0b7c3e9b221168bcd92497cfcafe5e546ca68501dfc51f5d012f947",
"bytes": 7668131810,
"url": null
},
{
"path": "umt5-base-tokenizer/special_tokens_map.json",
"sha256": "14ef492a69e59396e918521a5f8394af1acf9f8dddd47273693b2b42570dfd12",
"bytes": 6623,
"url": null
},
{
"path": "umt5-base-tokenizer/spiece.model",
"sha256": "e3909a67b780650b35cf529ac782ad2b6b26e6d1f849d3fbb6a872905f452458",
"bytes": 4548313,
"url": null
},
{
"path": "umt5-base-tokenizer/tokenizer.json",
"sha256": "af904105ce1071b1202bba0059a841f4a7b85b48b6ec179c4948e3483476e0dd",
"bytes": 16853013,
"url": null
},
{
"path": "umt5-base-tokenizer/tokenizer_config.json",
"sha256": "616015e2b6c26f9a98ae3171634518d440a46260ebc1115c2596792d33827bdc",
"bytes": 6841,
"url": null
}
]
},
"soprano": {
"source": "mirror",
"dest": "comfyui",
"destPath": "soprano/Soprano-1.1-80M",
"repoPath": "engines/soprano",
"junction": {
"link": "ComfyUI\\models\\soprano"
},
"gate": null,
"origin": {
"repo": "ekwek/Soprano-1.1-80M",
"url": "https://huggingface.co/ekwek/Soprano-1.1-80M"
},
"licence": {
"weights": "Apache-2.0",
"verified": "wheel LICENSE + HF cardData (v25.52.0 addendum)"
},
"postInstall": [
"ParrotSopranoTTS node in parrot-tts-nodes vendors the soprano 0.2.0 SDK; ZERO pip installs needed",
"junctioned into ComfyUI\\models\\soprano — the FireRed/VoxCPM 'no searching branch' shape"
],
"notes": "localPath resolved and verified non-empty 2026-08-18 (config.json, decoder.pth, model.safetensors, tokenizer.json, generation_config.json, special_tokens_map.json, tokenizer_config.json). Also matches the HF hub cache dir models--ekwek--Soprano-1.1-80M on this machine, corroborating the repo id.",
"files": [
{
"path": "config.json",
"sha256": "b4e487a7a3bbee21fa0fc3c0faaaf404911e5f843058b3a10d1102170d4c2bb5",
"bytes": 1118,
"url": null
},
{
"path": "decoder.pth",
"sha256": "11245487e4588b0f4128d6b4c82035212bad42ffb26f36e059e9c7f861c1f8c4",
"bytes": 121448447,
"url": null
},
{
"path": "generation_config.json",
"sha256": "83e8d44a0cfae1f40801b6f96ba8090286cfd5f707ae7fa3ec64e6f68550828c",
"bytes": 111,
"url": null
},
{
"path": "model.safetensors",
"sha256": "10e8c4223e14115688a1c8a1c5464ebb78826f17fb98ad459da9bce5471866ed",
"bytes": 159449112,
"url": null
},
{
"path": "special_tokens_map.json",
"sha256": "c68f391a753fa4ff6df6890a62fd584eba54ee5d1773c5f388e80f521ebba942",
"bytes": 28,
"url": null
},
{
"path": "tokenizer.json",
"sha256": "eb27ca6c9b55d2f880659a9619d6b8f9b9e6d8553d3ea424f5427788688d7286",
"bytes": 1630675,
"url": null
},
{
"path": "tokenizer_config.json",
"sha256": "06e83fdb0bdf94eef83ebb08f30620c8ab640344babb4d405ff27befd1a4c270",
"bytes": 1366802,
"url": null
}
]
},
"gptsovits": {
"source": "mirror",
"dest": "runtime",
"destPath": "GPT_SoVITS/pretrained_models",
"repoPath": "engines/gptsovits",
"junction": {
"link": "(machine-local path — not published; see the private engines.json)"
},
"gate": null,
"origin": {
"repo": "GPT-SoVITS official v2ProPlus prezipped package (RVC-Boss/GPT-SoVITS)",
"url": "https://huggingface.co/lj1995/GPT-SoVITS"
},
"licence": {
"weights": "MIT",
"verified": "TTS ENGINE REPORT.md §11: 'MIT / MIT'"
},
"postInstall": [
"direct route (not a ComfyUI workflow) — server is functions\\tts-and-sts-engines\\gptsovits-runtime\\...\\api_v2.py, auto-spawned by main.js",
"(machine-local path — not published; see the private engines.json) junction exists specifically to dodge a MAX_PATH failure in the runtime's compiled sklearn extensions (deep repo paths break it) — the junction is machine-local and reversible, files stay in-repo"
],
"notes": "localPath resolved via the (machine-local path — not published; see the private engines.json) junction and verified non-empty 2026-08-18 (chinese-hubert-base\\, chinese-roberta-wwm-ext-large\\, gsv-v2final-pretrained\\, gsv-v4-pretrained\\, s1bert25hz-...ckpt, s1v3.ckpt, s2D488k.pth, s2G488k.pth, s2Gv3.pth, sv\\, v2Pro\\).",
"files": [
{
"path": ".gitignore",
"sha256": "5c89a8a9c5af258d02d0bfd3582c75963bea53baa1f5f5132662b80020a63452",
"bytes": 13,
"url": null
},
{
"path": "chinese-hubert-base/config.json",
"sha256": "c3e5060a1277e0f078cc6be9da4528a605dba6ece93018981fe2c820e5c7b103",
"bytes": 1449,
"url": null
},
{
"path": "chinese-hubert-base/preprocessor_config.json",
"sha256": "dcd684124d06722947939d41ea6ae58dbf10968c60a11a29f23ddc602c64a29b",
"bytes": 212,
"url": null
},
{
"path": "chinese-hubert-base/pytorch_model.bin",
"sha256": "24164f129c66499d1346e2aa55f183250c223161ec2770c0da3d3b08cf432d3c",
"bytes": 188811417,
"url": null
},
{
"path": "chinese-roberta-wwm-ext-large/config.json",
"sha256": "3d57de2fd7e80d0e5c8ff194f0bbb6baa10df7e43fc262a0cc71298a78b0a3e5",
"bytes": 963,
"url": null
},
{
"path": "chinese-roberta-wwm-ext-large/pytorch_model.bin",
"sha256": "e53a693acc59ace251d143d068096ae0d7b79e4b1b503fa84c9dcf576448c1d8",
"bytes": 651225145,
"url": null
},
{
"path": "chinese-roberta-wwm-ext-large/tokenizer.json",
"sha256": "173796956820ea27bd14f76bf28162607ff4254807e2948253eb5b46f5bb643b",
"bytes": 268962,
"url": null
},
{
"path": "fast_langdetect/lid.176.bin",
"sha256": "7e69ec5451bc261cc7844e49e4792a85d7f09c06789ec800fc4a44aec362764e",
"bytes": 131266198,
"url": null
},
{
"path": "gsv-v2final-pretrained/s1bert25hz-5kh-longer-epoch=12-step=369668.ckpt",
"sha256": "732f94e63b148066e24c7f9d2637f3374083e637635f07fbdb695dee20ddbe1f",
"bytes": 155315150,
"url": null
},
{
"path": "gsv-v2final-pretrained/s2D2333k.pth",
"sha256": "8ae7fe8dd8c8f2e718de359e00edac88b0c71ab2fd10b07ad4cc45070eb8a836",
"bytes": 93534164,
"url": null
},
{
"path": "gsv-v2final-pretrained/s2G2333k.pth",
"sha256": "924fdccaa3c574bf139c25c9759aa1ed3b3f99e19a7c529ee996c2bc17663695",
"bytes": 106035259,
"url": null
},
{
"path": "gsv-v4-pretrained/s2Gv4.pth",
"sha256": "906fe22f48c3e037a389df291d4d32a9414e15dbb8f9628643e83aaced109ea4",
"bytes": 769025545,
"url": null
},
{
"path": "gsv-v4-pretrained/vocoder.pth",
"sha256": "4d611913df7b12d49e8976c944558d2d096816365edfc6c35a9e85b67dd14ed9",
"bytes": 57781109,
"url": null
},
{
"path": "models--nvidia--bigvgan_v2_24khz_100band_256x/bigvgan_generator.pt",
"sha256": "ee5e2f9cd60b51db75e1806f4fe7621733757586c541c78cb3dd369d5ba24476",
"bytes": 225179685,
"url": null
},
{
"path": "models--nvidia--bigvgan_v2_24khz_100band_256x/config.json",
"sha256": "d77e2c96583ca2296ac112a56ec7cc6bd5da4bf7681ceff18448bedc4fcf6512",
"bytes": 1402,
"url": null
},
{
"path": "s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt",
"sha256": "b1c1e17e9c99547a89388f72048cd6e1b41b5a18b170e86a46dfde0324d63eb1",
"bytes": 155093966,
"url": null
},
{
"path": "s1v3.ckpt",
"sha256": "87133414860ea14ff6620c483a3db5ed07b44be42e2c3fcdad65523a729a745a",
"bytes": 155284856,
"url": null
},
{
"path": "s2D488k.pth",
"sha256": "fc579c1db3c1e21b721001cf99d7a584214280df19b002e200b630a34fa06eb8",
"bytes": 93533667,
"url": null
},
{
"path": "s2G488k.pth",
"sha256": "020a014e1e01e550e510f2f61fae5e5f5b6aab40f15c22f1f12f724df507e835",
"bytes": 105973721,
"url": null
},
{
"path": "s2Gv3.pth",
"sha256": "f33abb1920076d988e1711d5f41b5c9c6d7f92575b4acf0ad4fae6a4ebf0cf19",
"bytes": 769035145,
"url": null
},
{
"path": "sv/pretrained_eres2netv2w24s4ep4.ckpt",
"sha256": "4f5a0bf73c61eb41b174e1bb54e7ee3c83233892be8e0af1f187024e8e581a35",
"bytes": 107528697,
"url": null
},
{
"path": "v2Pro/s2Dv2Pro.pth",
"sha256": "257a7f36ae49e9ff08790d2cd3d398592530852caba810197f8606e884a52c54",
"bytes": 126431971,
"url": null
},
{
"path": "v2Pro/s2Dv2ProPlus.pth",
"sha256": "635cd84bf6f7f9b8d41c88c7106f81d782c794c61f931845214ea037b0c5bef2",
"bytes": 126432639,
"url": null
},
{
"path": "v2Pro/s2Gv2Pro.pth",
"sha256": "0f8ead815234365edf045c6d86370ed6e4f440e8195be77ff0ea72684ad406a5",
"bytes": 162303657,
"url": null
},
{
"path": "v2Pro/s2Gv2ProPlus.pth",
"sha256": "d42a22bbbf65fb2bbdd45ad6a66841156977db45c7aabe0a6992ff378d9c7d3b",
"bytes": 200125741,
"url": null
}
]
},
"mossnano": {
"source": "mirror",
"dest": "comfyui",
"destPath": "moss-nano",
"repoPath": "engines/mossnano",
"junction": null,
"gate": null,
"origin": {
"repo": "OpenMOSS-Team/MOSS-TTS-Nano-100M-ONNX + OpenMOSS-Team/MOSS-Audio-Tokenizer-Nano-ONNX",
"url": "https://huggingface.co/OpenMOSS-Team/MOSS-TTS-Nano-100M-ONNX"
},
"licence": {
"weights": "Apache-2.0",
"verified": "'Apache 2.0 code AND weights' — mossnano-clone-v1/mossnano-preset-v1 manifest notes + parrot-tts-nodes VENDORED-README.md line 23"
},
"postInstall": [
"the parrot-tts-nodes ParrotMossNano node reads folder_paths.models_dir + 'moss-nano' directly — no junction needed, this sits natively inside ComfyUI's own models tree"
],
"notes": "localPath corrected by coordinator resolution (2026-08-18 follow-up): this machine's copy has NO engine installs of its own (kimi-tts-sts-suite is a parallel-work copy; settings.json still points comfyuiPath at the ORIGINAL tree's shared ComfyUI-Portable install), so mossnano's files were never going to be under the kimi tree — my first pass searched only there plus the original tree's generic models\\ root and missed ComfyUI's OWN models dir inside that install. Verified 2026-08-18: contains MOSS-Audio-Tokenizer-Nano-ONNX\\ (codec_browser_onnx_meta.json, moss_audio_tokenizer_decode_full.onnx, moss_audio_tokenizer_decode_shared.data, moss_audio_tokenizer_decode_step.onnx, moss_audio_tokenizer_encode.data, moss_audio_tokenizer_encode.onnx) and MOSS-TTS-Nano-100M-ONNX\\ (browser_poc_manifest.json, moss_tts_decode_step.onnx, moss_tts_global_shared.data, moss_tts_local_cached_step.onnx, moss_tts_local_decoder.onnx, moss_tts_local_fixed_sampled_frame.onnx, moss_tts_local_shared.data, moss_tts_prefill.onnx, tokenizer.model, tts_browser_onnx_meta.json) — the two ONNX repos, ~728MB total, exactly as the licence record describes. No Task 6 dependency notes needed — both repos are Apache-2.0 verified. WAVE E (v25.104.0, 2026-08-26): a second audio.cpp route exists on the SAME gate-installed bf16 GGUF - mossnano-tts-cpp, a text-only continuation ROLLER (no built-in voices in the CLI; the comfy preset's 18 voices are parrot-tts-nodes prompt codes, so the routes are deliberately NOT paired). Same Apache-2.0 weights, no new files.",
"files": [
{
"path": "MOSS-Audio-Tokenizer-Nano-ONNX/codec_browser_onnx_meta.json",
"sha256": "3e291c883bb7d11ff2fe8e964e3e495519760358859f35c951254c7741592731",
"bytes": 17036,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer-Nano-ONNX/moss_audio_tokenizer_decode_full.onnx",
"sha256": "0fbbafe3fd4afa2a019af5c5ced204af6e2d1db044fa40f021525d2aee95b4ac",
"bytes": 681902,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer-Nano-ONNX/moss_audio_tokenizer_decode_shared.data",
"sha256": "e69d52e0f4e84ca27850557ee54face46632d3a5a16c89bd246c7c408466dcad",
"bytes": 44198912,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer-Nano-ONNX/moss_audio_tokenizer_decode_step.onnx",
"sha256": "9527c86a29e1837edec1f74db57d5eeaadb3a715af3382703566460afed25855",
"bytes": 351400,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer-Nano-ONNX/moss_audio_tokenizer_encode.data",
"sha256": "aa751265b2bab2887eac224484546b194875aa7494b607115439b3dc6b228a2c",
"bytes": 44507136,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer-Nano-ONNX/moss_audio_tokenizer_encode.onnx",
"sha256": "eadea4a645abdcf98714c7aead122ee2ce7da6e080f9f80b977cd1ca8e19473a",
"bytes": 815775,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/browser_poc_manifest.json",
"sha256": "097d80e993dc29f0bae427590b4f77084a161cb578b50d82c29f455d5faa9eee",
"bytes": 503354,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/moss_tts_decode_step.onnx",
"sha256": "698cbc2fc1c2feca16e5895614ed52bbb32ded10f236c076f477b2e69abf32d8",
"bytes": 291483,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/moss_tts_global_shared.data",
"sha256": "bce8312c3df6a44545302cae229b61054fe0672e0b252ba59cba47adeed831dc",
"bytes": 440813568,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/moss_tts_local_cached_step.onnx",
"sha256": "aa9035fefc1c138a951a8bcfc0374fb03a25f1ece67f7f7f53bce349b84a1dd5",
"bytes": 53685,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/moss_tts_local_decoder.onnx",
"sha256": "51aa754301b38550a5f9adda0ad93bd3dc95819afb511e6dcabf4a90b345a454",
"bytes": 49231,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/moss_tts_local_fixed_sampled_frame.onnx",
"sha256": "40cdb00efc171c450cf91468e01429caa41b0252222cd308e978f58fe354afa8",
"bytes": 471262,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/moss_tts_local_shared.data",
"sha256": "bae7782032c0fb12490ab42afe009f87ae6c75a0f0596fc7b5c08e4d5ee93916",
"bytes": 229678080,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/moss_tts_prefill.onnx",
"sha256": "d56126dcd0574c2f15d98fc6b35eda68d0386b5bd9c5e38e28548d6f2ea8f3db",
"bytes": 283305,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/tokenizer.model",
"sha256": "c353ee1479b536bf414c1b247f5542b6607fb8ae91320e5af1781fee200fddff",
"bytes": 470897,
"url": null
},
{
"path": "MOSS-TTS-Nano-100M-ONNX/tts_browser_onnx_meta.json",
"sha256": "3edf25232dcd0af3d061c837e9a968a39e2f8592e06777d740503c4f2244f95c",
"bytes": 4487,
"url": null
}
]
},
"kokoro": {
"source": "mirror",
"dest": "pack",
"destPath": ".",
"repoPath": "engines/kokoro",
"junction": null,
"gate": null,
"origin": {
"repo": "onnx-community/Kokoro-82M-v1.0-ONNX",
"url": "https://huggingface.co/onnx-community/Kokoro-82M-v1.0-ONNX"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF cardData lineage via onnx-community, recorded in kokoro-tts-v1 manifest notes (v25.34.0)"
},
"postInstall": [
"pack: stavsap/comfyui-kokoro + PARROT phonemizer shim (see CLAUDE.md env maintenance)"
],
"notes": "model + voices bin live in the pack dir under the ORIGINAL tree's shared ComfyUI install (this kimi copy has no engine installs of its own — settings.json still points comfyuiPath at the original tree). localPath resolved and verified non-empty 2026-08-18 (kokoro_v1.onnx, voices_v1.bin present directly in the pack root).",
"files": [
{
"path": ".github/workflows/publish.yml",
"sha256": "8d2bf3990fe7dcf771ec44498e6b2e3c8e760a2e358fa326337b123d70346bb3",
"bytes": 671,
"url": null
},
{
"path": ".gitignore",
"sha256": "2ed2b065caf5793010b8653e64fe63523b06ccd4ccaf5bc1ea05e0b01fde4d6b",
"bytes": 3629,
"url": null
},
{
"path": ".meta/generator.png",
"sha256": "16e535bb0185d32340461dddf1f96de846ad646a374c2efd52e3d93f0bd6ca1c",
"bytes": 39833,
"url": null
},
{
"path": ".meta/img.png",
"sha256": "d746119ca562ba35cc36bc1900c4a7109789200b6bb7ce59605901c207df39b3",
"bytes": 114075,
"url": null
},
{
"path": ".meta/lipsync.png",
"sha256": "bdebab6196c825d11b55c2b99c176899223892f2784aaa6fc12eb94ea0c94f06",
"bytes": 366816,
"url": null
},
{
"path": ".meta/speaker.png",
"sha256": "0e470852e8c57707b009125f23822f18b92bde2eb11f7085ee324521956f472f",
"bytes": 22946,
"url": null
},
{
"path": ".meta/speaker_combiner.png",
"sha256": "1b3be0d512954181c03217326694a55ad885b9f69f15a2eabb2cdb562c9b1280",
"bytes": 27748,
"url": null
},
{
"path": ".meta/workflow.png",
"sha256": "dc1750c4e60d7cf13e421abfdec1faca447f6b121539b328acff57043b83dd5a",
"bytes": 125822,
"url": null
},
{
"path": ".meta/yt_img.png",
"sha256": "f32f0c8a45bcb8a9d6786983a3624c1b145dcf58f8b71812e4b25da511d4124e",
"bytes": 296685,
"url": null
},
{
"path": "CHANGE_LOG.md",
"sha256": "609df8e6c69fc3ceef7d6b1011a5c97b1e0e2781c94b89e2756809a6b7917387",
"bytes": 291,
"url": null
},
{
"path": "ComfyUIKokoro.py",
"sha256": "85eb4f58483f8d7a93b1942879bdd6360a7b1677ff9949d4fdc3e03e72331b74",
"bytes": 8502,
"url": null
},
{
"path": "LICENSE",
"sha256": "1eb85fc97224598dad1852b5d6483bbcf0aa8608790dcc657a5a2a761ae9c8c6",
"bytes": 11558,
"url": null
},
{
"path": "README.md",
"sha256": "301c282635e1dea54c496093e547967c8eff708eba87b108e1a28262eeaddaba",
"bytes": 2739,
"url": null
},
{
"path": "__init__.py",
"sha256": "ff27dccf0ff8ab39da120a49eab405d4738dd0646471bf934583df080b7eaa4b",
"bytes": 1353,
"url": null
},
{
"path": "__pycache__/ComfyUIKokoro.cpython-312.pyc",
"sha256": "80c3067d98c61e8f85c5bd6076c057872e7a238f6b52f6535a4b2b3a9d18bbee",
"bytes": 11339,
"url": null
},
{
"path": "__pycache__/__init__.cpython-312.pyc",
"sha256": "ee54788d9e33086d4523b7bc5c2e5d8e99843012640e4d43447ff4a4cfd669d0",
"bytes": 829,
"url": null
},
{
"path": "kokoro_v1.onnx",
"sha256": "7d5df8ecf7d4b1878015a32686053fd0eebe2bc377234608764cc0ef3636a6c5",
"bytes": 325532387,
"url": null
},
{
"path": "pyproject.toml",
"sha256": "f6d7965b312e58451417df0312a3f966652328438487f04fe86a0e915b48f569",
"bytes": 491,
"url": null
},
{
"path": "requirements.txt",
"sha256": "2297ce19a20d21d4a3d0f166a8cd7c5741ba680f0ca1db31b874d75dfc71e5f4",
"bytes": 49,
"url": null
},
{
"path": "voices_v1.bin",
"sha256": "a6d439ed196348d1a3529619db4939487619e6155c726a5fb0fd00538d373385",
"bytes": 28214398,
"url": null
}
]
},
"neutts": {
"source": "mirror",
"dest": "hub",
"repoPath": "engines/neutts",
"junction": null,
"gate": null,
"origin": {
"repo": "neuphonic/neutts-air-q8-gguf + neuphonic/neucodec",
"url": "https://huggingface.co/neuphonic/neutts-air-q8-gguf"
},
"licence": {
"weights": "Apache-2.0",
"verified": "SDK + Air weights cardData (v25.53.0 addendum); Nano sibling EXCLUDED (revenue-capped licence, never shipped)"
},
"postInstall": [
"ParrotNeuTTS node in parrot-tts-nodes vendors neutts 1.4.1 + neucodec 0.0.6",
"codec's THIRD hidden dependency facebook/w2v-bert-2.0 is NOT licence-verified — deliberately left OUT of localPath parts, see notes"
],
"notes": "Array-of-parts row: neutts-air-q8-gguf (766MB GGUF backbone) and neucodec (1.1GB torch codec) are two separate HF repos that must both ship under engines/neutts/. Both snapshot dirs resolved and verified non-empty 2026-08-18 (neutts-air-Q8_0.gguf; config.json + meta.yaml + pytorch_model.bin). NOT included: the codec's own hidden second repo facebook/w2v-bert-2.0 (fetched at first codec load, per CLAUDE.md 'the Echo lesson'). Task 6 (2026-08-18) follow-up: facebook/w2v-bert-2.0's licence is now checked (HF API cardData: MIT, permissive) — it is still deliberately left OUT of localPath as a THIRD part, because it is a runtime dependency fetched on demand by the codec rather than something the neutts engine's own manifest cites directly; a future task can add it as a third array part if bundling it up front is wanted, now that its licence is a known quantity rather than an open question. v25.85.0 (audio.cpp exclusives wave 4): the SAME `neutts` kind now also carries the audio.cpp workflow neutts2e-preset-cpp, which uses a DIFFERENT file NOT part of this mirror row - NeuTTS-2E neutts-2e-orig.gguf from audio-cpp/audio.cpp-gguf (3,016,181,288 bytes, sha256 fb1b437f6ae9ba005930a38977573f994bc08290f3b16ee6ca99028a2aeb402d), whose HF card (neuphonic/neutts-2e, gated) says license 'other' = the NeuTTS Open License (revenue-capped, the same family as the Nano sibling this row already excludes); the audio-cpp card row's 'Apache-2.0' for it is a MISLABEL. LOCAL USE only; never mirrored. The mirror-eligible parts above stay exactly the Apache-2.0 Air files.",
"destPath": {
"neutts-air-q8-gguf": "models--neuphonic--neutts-air-q8-gguf/snapshots/3c0f88293e3533ca0168905e75ef03be1c5aa906",
"neucodec": "models--neuphonic--neucodec/snapshots/30c1fdd19e68aee65d542cf043750d4c0165893e"
},
"files": [
{
"path": "neutts-air-q8-gguf/neutts-air-Q8_0.gguf",
"sha256": "2f127bd6741704ccfe015dc9319254fc751dd146766c2bab17214bc036e73fd7",
"bytes": 802658528,
"url": null
},
{
"path": "neucodec/config.json",
"sha256": "944629de35ef9dcdf38746b8041e4ea3d7b781919f4b55d98405f78b0338d2ff",
"bytes": 2958,
"url": null
},
{
"path": "neucodec/meta.yaml",
"sha256": "232631816b83ee6c8929b7f0b960aedb7b215caa73dd96f7a0e60946b2a1319b",
"bytes": 37,
"url": null
},
{
"path": "neucodec/pytorch_model.bin",
"sha256": "30c3ea13ceeb2de693c56e5e33a1b7e00d44c95dcdd08a4ed0d552d0bf59ebdf",
"bytes": 1160509432,
"url": null
}
]
},
"pocket-preset": {
"source": "mirror",
"dest": "hub",
"repoPath": "engines/pocket-preset",
"junction": null,
"gate": null,
"origin": {
"repo": "kyutai/pocket-tts-without-voice-cloning",
"url": "https://huggingface.co/kyutai/pocket-tts-without-voice-cloning"
},
"licence": {
"weights": "CC-BY-4.0",
"verified": "cardData both repos (v25.42.0 addendum); ATTRIBUTION required in README"
},
"postInstall": [
"ParrotPocketTTS node, pocket-preset-v1 manifest — the app kind is 'pocket' (this row is the preset HALF only; the clone half is the separate gated row pocket-clone)",
"26 English voice embeddings, ~6MB each, all prefetched at the SDK's pinned rev"
],
"notes": "single pinned snapshot e041936c (26 files); the stale partial d29db797 snapshot was dropped 2026-08-18. Distinct from the gated kyutai/pocket-tts clone repo and from kyutai/mimi (the codec dependency). Task 6 (2026-08-18) follow-up: kyutai/mimi's licence is now checked (HF API cardData: CC-BY-4.0, permissive) — still deliberately left OUT of localPath, same reasoning as neutts's facebook/w2v-bert-2.0 dependency above: it's a runtime dependency fetched on demand, not something the pocket-preset manifest itself cites, and can be added as its own part later now that its licence is known rather than open.",
"destPath": {
"pocket-tts-without-voice-cloning": "models--kyutai--pocket-tts-without-voice-cloning/snapshots/e041936c75475d350b405bc870bcf7c22da4e9e6"
},
"files": [
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/alba.safetensors",
"sha256": "69c32db63ca56843d994f81f343f62e0bf2d73f7e4c9bc73e44bb1110b1d8845",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/anna.safetensors",
"sha256": "5ea82f78db006c9fd34e32ddd5aae82674b5b32646097977436458d00af80dfa",
"bytes": 7816440,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/azelma.safetensors",
"sha256": "9f3e69f29075f991fd47774566865ef0e0e637cb5a35992c9919761b5b84b1de",
"bytes": 7963896,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/bill_boerst.safetensors",
"sha256": "75610127d44e0b05b442154f80f89f993df235aecc6cad7070f11000d006c188",
"bytes": 6735096,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/caro_davy.safetensors",
"sha256": "a5961b63a2e7a5cfd7edc383aa9042fb70fd14a9dee6310cdc633881a7f2449a",
"bytes": 5260536,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/charles.safetensors",
"sha256": "299edc20182eeccfbf94e308626f259da4fbf339daa8d5905218f2b1774639b8",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/cosette.safetensors",
"sha256": "c4fdc15f5a3a20c44dd0064a37e87d15d25562936e8dbad7e07b9832015a545d",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/eponine.safetensors",
"sha256": "bda3b76a384ff355fe0350736387765946304ae8ca16e59f60ea3296a1c99cc6",
"bytes": 6931704,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/estelle.safetensors",
"sha256": "ccebef7f51762c7fc08870f5ecf268e8713551ae2c9f7984ddaec0c1e1c77153",
"bytes": 8258808,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/eve.safetensors",
"sha256": "ea9c2faf862a6c9d2cb61910fdf02842ae56940382cc8c1000fdb1b43269692b",
"bytes": 6538488,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/fantine.safetensors",
"sha256": "51a8a4355d7f912d4959e4b1918314fda85ad47eba0a33a1d78a4a505d3465f5",
"bytes": 6538488,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/george.safetensors",
"sha256": "0c1c6c57c55a98d81254b33728150c7776f40647fe95258d1a6c1a02780b5d02",
"bytes": 6243576,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/giovanni.safetensors",
"sha256": "a5ee718157ec1c6fd9c1e66a7733c7e6337d474a74a7da756455d27717885c59",
"bytes": 4621552,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/jane.safetensors",
"sha256": "37386227ca8ec5bf1b8e516c13d132ce5ff5437a304fe90129a1c62f41d9a008",
"bytes": 7374072,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/javert.safetensors",
"sha256": "0ae88e03ca4e76a0e16cbf321a807428febda9d9e9bc0358c02e7f9c9e2c263b",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/jean.safetensors",
"sha256": "90be4b8f50bb4d2dbe27e3fb4e31417cf6a57928931f0a60426a1748821a3d12",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/juergen.safetensors",
"sha256": "e74d67b38339fc01118e3bbe2c90d6d40e601b161dbf7be906421956ba80a532",
"bytes": 6243576,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/lola.safetensors",
"sha256": "34972e86b07b17272a8061460054a08119399cace9454979052b9e2d96664e8c",
"bytes": 5948664,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/marius.safetensors",
"sha256": "04f84efcb77a0547ba582c058db496f7ff4920891d49d37b9950d128422582a8",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/mary.safetensors",
"sha256": "a8f2adf260cab966fe0a113d6b549d6efdeaa79de544ae0ff34b5b6a41445a59",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/michael.safetensors",
"sha256": "8937f724ac4719b9aa51ea0ba1f18f9de0af7a663ad6263558266c1a53c9722d",
"bytes": 7275768,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/paul.safetensors",
"sha256": "ed7a019168f94dfe77009f1b0de59387abc6fbb0db954d38ce722ecb77da61aa",
"bytes": 6980856,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/peter_yearsley.safetensors",
"sha256": "dd977a6e15591e347c9a23fa7cc09e35a65b462917f5eeb162baff6dc9e3f685",
"bytes": 3736816,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/rafael.safetensors",
"sha256": "ac19f099f6cd839875a629c3e2e91e0dfc2c197acf2875db168d0ae244fb58bd",
"bytes": 6194424,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/stuart_bell.safetensors",
"sha256": "5a49da7ca5df05d02587ec4a0981c0d318e045f68e24423c4203ce474d9b33dc",
"bytes": 5260536,
"url": null
},
{
"path": "pocket-tts-without-voice-cloning/languages/english/embeddings/vera.safetensors",
"sha256": "4bf50ddd957b5d218b264fdcf18efbbc7384d12da3eca98ca19b9e8dd6976acc",
"bytes": 6735096,
"url": null
}
]
},
"pocket-clone": {
"source": "gated",
"dest": null,
"repoPath": "engines/pocket-clone",
"junction": null,
"gate": {
"repo": "kyutai/pocket-tts",
"why": "HF-gated repo — the clone half of PocketTTS; POLICY: never re-hosted, the owner's own accepted gate is required per machine"
},
"origin": {
"repo": "kyutai/pocket-tts",
"url": "https://huggingface.co/kyutai/pocket-tts"
},
"licence": {
"weights": "CC-BY-4.0",
"verified": "gate kyutai/pocket-tts — POLICY: never re-hosted"
},
"postInstall": [
"app kind is 'pocket' (clone half); the reference-audio workflow is pocket-clone-v1"
],
"notes": "Gated — deliberately never mirrored regardless of what is present in the local HF hub cache. Distribution consists solely of pointing the owner at the gate URL.",
"files": []
},
"orpheus": {
"source": "gated",
"dest": null,
"repoPath": "engines/orpheus",
"junction": null,
"gate": {
"repo": "canopylabs/orpheus-3b-0.1-ft",
"why": "HF-gated (fine-grained token, 'read gated repos' permission); gate-state discrepancy RESOLVED 2026-08-18, see notes"
},
"origin": {
"repo": "canopylabs/orpheus-3b-0.1-ft",
"url": "https://huggingface.co/canopylabs/orpheus-3b-0.1-ft"
},
"licence": {
"weights": "Apache-2.0",
"verified": "gate canopylabs/orpheus-3b-0.1-ft; gate-state discrepancy resolved (Task 6) — CLAUDE.md was accurate, the workflow manifest note was stale"
},
"postInstall": [
"orpheus-tts-v1 manifest — the loaded 3B rides ComfyUI's own node-output cache; snac must be >=1.2.1 (pack pins 0.1.0, re-apply by symptom)"
],
"notes": "Task 6 (2026-08-18): gate-state discrepancy RESOLVED without touching any token. Checked the local HF hub cache directory (never the token itself, per the task's own constraint): (machine-local path — not published; see the private engines.json) contains FOUR real safetensors shards, not stubs — model-00001-of-00004.safetensors (4,948,557,560 bytes), -00002 (4,932,808,960), -00003 (3,322,130,968), -00004 (1,928,478,848), summing to 15,131,976,336 bytes = 14.09 GiB, which matches CLAUDE.md's 'HF-gated weights flow' record ('orpheus accepted') and the roster row's '14.1GB FP32 shards' almost exactly. VERDICT: CLAUDE.md is ACCURATE (gate accepted, weights prefetched); the orpheus-tts-v1 workflow manifest's own note — 'HF-GATED (\\'auto\\'): the OWNER must accept the terms ... NOT prefetched for exactly that reason' — is STALE (the manifest was written before the owner accepted the gate and fetched the weights, and nothing has updated that note since). Per this task's constraints the workflow manifest was NOT edited — only read, and only this engines.json row records the correction. Gated, non-graduable by policy regardless of licence (Apache-2.0) — never re-hosted; the owner's own accepted gate is required per machine.",
"files": []
},
"sesame": {
"source": "gated",
"dest": null,
"repoPath": "engines/sesame",
"junction": null,
"gate": {
"repo": "sesame/csm-1b",
"why": "HF-gated (fine-grained token, 'read gated repos' permission)"
},
"origin": {
"repo": "sesame/csm-1b",
"url": "https://huggingface.co/sesame/csm-1b"
},
"licence": {
"weights": "Apache-2.0",
"verified": "Apache-2.0 per HF card"
},
"postInstall": [
"sesame-csm-v1 manifest — context-conditioned cloner, zero new pip packages; Llama tokenizer rides the ungated unsloth mirror"
],
"notes": "Gated. ckpt.pt is the gated artifact (6.2GB).",
"files": []
},
"silero": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/silero",
"junction": null,
"gate": null,
"origin": {
"repo": "models.silero.ai (plain HTTP)",
"url": "https://models.silero.ai/models/tts/en/v3_en.pt"
},
"licence": {
"weights": "CC BY-NC-SA 4.0",
"verified": "silero-tts-v1 manifest notes + parrot-tts-nodes README L21 (bold NON-COMMERCIAL)"
},
"postInstall": [
"sha256 02b71034d9f13bc4001195017bac9db1c6bb6115e03fea52983e8abcff13b665, 57,194,546 bytes — install.bat verifies"
],
"notes": "UN-HOSTABLE: non-commercial weights. Pointer README only.",
"files": []
},
"spark": {
"source": "origin",
"dest": "comfyui",
"repoPath": "engines/spark",
"junction": {
"link": "ComfyUI\\models\\TTS\\SparkTTS\\Spark-TTS-0.5B"
},
"gate": null,
"origin": {
"repo": "SparkAudio/Spark-TTS-0.5B",
"url": "https://huggingface.co/SparkAudio/Spark-TTS-0.5B"
},
"licence": {
"weights": "CC-BY-NC-SA 4.0",
"verified": "relicensed at HF commit b63203d4; pack header (1038lab/Comfyui-Spark-TTS) STALE/WRONG, still claims Apache-2.0"
},
"postInstall": [
"SparkTTSCore resolves folder_paths.models_dir/TTS/SparkTTS with NO searching branch — must be a junction, never a plain download into a wrong path"
],
"notes": "UN-HOSTABLE: non-commercial weights (relicensed from Apache after training-data licensing concerns) — never mirror. Own weights already present locally at models\\comfyui\\TTS\\SparkTTS\\Spark-TTS-0.5B on this machine (owner-fetched directly from origin, not something this component re-distributes).",
"files": []
},
"echo": {
"source": "origin",
"dest": "app-models",
"repoPath": "engines/echo",
"junction": null,
"gate": null,
"origin": {
"repo": "jordand/echo-tts-base + jordand/fish-s1-dac-min",
"url": "https://huggingface.co/jordand/echo-tts-base"
},
"licence": {
"weights": "CC-BY-NC-SA-4.0",
"verified": "HF API cardData, both repos (jordand/echo-tts-base, jordand/fish-s1-dac-min), fetched 2026-08-18"
},
"postInstall": [
"echo-tts-clone-v1 manifest — needs TWO repos: echo-tts-base (5.6GB) AND fish-s1-dac-min (1.87GB, the audio autoencoder), both moved into the shared models root"
],
"notes": "Task 6 (2026-08-18): CONFIRMED as expected (brief step 2) — CC-BY-NC-SA-4.0 on both repos. Stays origin, non-commercial, fails the graduation bar regardless of file availability.",
"files": []
},
"higgs-v3": {
"source": "origin",
"dest": "app-models",
"repoPath": "engines/higgs-v3",
"junction": null,
"gate": null,
"origin": {
"repo": "bosonai/higgs-audio-v3-tts-4b",
"url": "https://huggingface.co/bosonai/higgs-audio-v3-tts-4b"
},
"licence": {
"weights": "BOSON HIGGS TTS 3 RESEARCH AND NON-COMMERCIAL LICENSE AGREEMENT",
"verified": "local LICENSE file read directly, 2026-08-18, at models\\comfyui\\TTS\\higgs_audio_v3\\higgs-audio-v3-tts-4b\\LICENSE (ships on disk with the weights)"
},
"postInstall": [
"higgs-v3-clone-v1 manifest — ~11GB, moved deliberately (v24.96) under models\\comfyui\\TTS\\higgs_audio_v3\\higgs-audio-v3-tts-4b so the plain model name can be found (no searching branch otherwise)"
],
"notes": "Task 6 (2026-08-18): CONFIRMED as expected (brief step 2) — the shipped LICENSE file transcribed verbatim. Title: 'BOSON HIGGS TTS 3 RESEARCH AND NON-COMMERCIAL LICENSE AGREEMENT', dated 2026-07-08. Section I states it is 'a research and non-commercial source-available model license and is NOT an open source license'; Section II-A grants a narrow free 'Creator Use Grant' (podcasts/videos/audiobooks with an attribution requirement) but explicitly EXCLUDES 'hosting, serving, or otherwise making the Higgs Materials available to others' and 'redistributing, reselling, sublicensing, or otherwise distributing the Higgs Materials' without a separate written commercial licence from Boson (Section II-A(c)(i)-(ii)). This directly forbids the kind of redistribution a mirror repo performs. Stays origin; not eligible for graduation on this licence under any reading. WAVE C (v25.102.0, 2026-08-26): audio.cpp GGUF twins exist (higgs-clone-cpp + the higgs-tts-cpp text-only roller; TTS 4B q8_0 from audio-cpp/audio.cpp-gguf, sha-locked wave 11). The SAME Boson-NC-licensed weights repackaged - the licence above carries; LOCAL use only, never mirrored.",
"files": []
},
"supertonic": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/supertonic",
"junction": null,
"gate": null,
"origin": {
"repo": "Supertone/supertonic-3",
"url": "https://huggingface.co/Supertone/supertonic-3"
},
"licence": {
"weights": "BigScience Open RAIL-M License (dated 2022-08-18)",
"verified": "HF API cardData Supertone/supertonic-3 (license: \"openrail\") + the repo's own raw LICENSE file at huggingface.co/Supertone/supertonic-3/raw/main/LICENSE, both fetched 2026-08-18"
},
"postInstall": [
"Anonymzx/ComfyUI-Supertonic3TTS pack (MIT) over the official supertonic 1.3.1 SDK (MIT); weights prefetched into the pack's own models\\supertonic-3 dir, ~385MB"
],
"notes": "Task 6 (2026-08-18): CONFIRMED as expected (brief step 2) — the exact variant is the BigScience Open RAIL-M License (the general-purpose successor to the original BLOOM-specific RAIL license; use-based restrictions apply, not a pure permissive licence). Link for the owner to read in full: https://huggingface.co/Supertone/supertonic-3/raw/main/LICENSE . UN-HOSTABLE: usage-restricted weights — never mirror without reading the terms first. Bucket/source left UNCHANGED per instructions (owner decision pending, not this task's call).",
"files": []
},
"dramabox": {
"source": "origin",
"dest": "app-models",
"repoPath": "engines/dramabox",
"junction": null,
"gate": null,
"origin": {
"repo": "ResembleAI/Dramabox + Lightricks/gemma-3-12b-it-qat-q4_0-unquantized",
"url": "https://huggingface.co/ResembleAI/Dramabox"
},
"licence": {
"weights": "LTX-2 Community (conditional)",
"verified": "dramabox-clone-v1 manifest notes — check terms before any commercial use"
},
"postInstall": [
"16.36GB total under models\\comfyui\\TTS\\dramabox\\DramaBox: 8.6GB DramaBox repo + 7.8GB Gemma-3-12B 4-bit text encoder",
"needed the vendored-xformers Blackwell fallback patch — see CLAUDE.md engine environment maintenance"
],
"notes": "conditional community licence — treat as UN-HOSTABLE without re-reading LTX-2 Community terms; never mirror on assumption alone. WAVE D (v25.103.0, 2026-08-26): audio.cpp GGUF twins exist (dramabox-clone-cpp + the dramabox-tts-cpp text-only roller; the 18.9 GB q8_0 bundle from audio-cpp/audio.cpp-gguf, sha-locked wave 12 - the largest file on that route). The SAME conditionally-licensed LTX-2 Community weights repackaged - the licence above carries; LOCAL use only, never mirrored.",
"files": []
},
"piper": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/piper",
"junction": null,
"gate": null,
"origin": {
"repo": "rhasspy/piper-voices (per-voice repos; not explicitly cited in the manifest — best available)",
"url": "https://huggingface.co/rhasspy/piper-voices"
},
"licence": {
"weights": "PER-VOICE, VARIES",
"verified": "piper voice param description + TTS ENGINE REPORT.md L980-982; runtime piper-tts 1.6.1 is GPL-3.0"
},
"postInstall": [
"yuvraj108c/ComfyUI-PiperTTS pack + a THREE-EDIT PARROT FORK (RETURN_TYPES STRING->AUDIO, synthesize->synthesize_wav, wave.open 'w'->'wb') plus the 'best' quality resolver fork",
"82 best-quality voices prefetched, 5.0GB — the catalog's own config size table is STALE, verify by PARSING not size"
],
"notes": "the manifest does not cite an explicit HF repo id for the per-voice weights — origin.url points at rhasspy/piper-voices as the ecosystem-standard host; confirm the exact source(s) actually used in Task 6. licence.weights is per-voice by design (each of the 82 voices carries its own licence, see VOICES.md).",
"files": []
},
"chatterbox": {
"source": "mirror",
"dest": "comfyui",
"repoPath": "engines/chatterbox",
"junction": null,
"gate": null,
"origin": {
"repo": "ResembleAI/chatterbox",
"url": "https://huggingface.co/ResembleAI/chatterbox"
},
"licence": {
"weights": "MIT",
"verified": "HF API cardData ResembleAI/chatterbox, fetched 2026-08-18"
},
"postInstall": [
"covers chatterbox-clone-v1, chatterbox-23lang-v1 and chatterbox-sts-v1 — one distribution unit for the whole 'chatterbox' kind"
],
"notes": "Task 6 (2026-08-18): MIT confirmed via HF API cardData — the manifests' own 'licence NOT verified from the repo' prose is now stale text inside those manifest files (out of scope to edit; this row is the authority). Graduated to mirror: both localPath parts verified non-empty on this machine (models\\TTS\\chatterbox\\English ~13GB — s3gen/ve/t3_cfg per chatterbox-clone-v1's own MODELS note; models\\TTS\\chatterbox_official_23lang\\Official 23-Lang ~13GB per chatterbox-23lang-v1's own MODELS note). A third local variant, chatterbox_turbo (t3_turbo_v1.safetensors etc.), sits alongside on disk but is NOT included here — its upstream repo id was not identified this pass and it is not cited by any of the three chatterbox manifests; leave for a future task rather than guess. v25.88.0 (the STS lane): the SAME kind now also carries the audio.cpp workflow chatterbox-vc-cpp (the voice-changer twin of chatterbox-sts-v1), which uses a DIFFERENT file NOT part of this mirror row - Chatterbox-GGUF/chatterbox-f16.gguf from audio-cpp/audio.cpp-gguf (3,744,360,386 bytes, sha256 b872f908605f5ccbdd8eaa9e9c92a521c602b70cbec635d46e86eb70a3f8e713; MIT, the same ResembleAI weights in GGUF form, fetched by fetch.js wave 6); local use, not redistributed by us. WAVE E (v25.104.0, 2026-08-26): a second audio.cpp route exists on the SAME wave-6 f16 GGUF - chatterbox-clone-cpp (--task clon), the TTS twin of the comfy chatterbox-clone, beside the v25.88.0 voice changer. Same MIT weights, no new files.",
"destPath": {
"english": "TTS/chatterbox/English",
"official-23lang": "TTS/chatterbox_official_23lang/Official 23-Lang"
},
"files": [
{
"path": "english/.gitattributes",
"sha256": "11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361",
"bytes": 1519,
"url": null
},
{
"path": "english/Cangjie5_TC.json",
"sha256": "7073fd9de919443ae88e0bd2449917a65fe54898a4413ed1edcc4b67f28bce8c",
"bytes": 1920163,
"url": null
},
{
"path": "english/README.md",
"sha256": "6ac86d5ec7e85a9cf6566a34c6aae60ac1d50a6fceccade4c2aa57cfa6f590b4",
"bytes": 10081,
"url": null
},
{
"path": "english/conds.pt",
"sha256": "6552d70568833628ba019c6b03459e77fe71ca197d5c560cef9411bee9d87f4e",
"bytes": 107374,
"url": null
},
{
"path": "english/grapheme_mtl_merged_expanded_v1.json",
"sha256": "69632f47220a788a52ce2661d096453c5655e9bf25289d89a8d832c46ee07dbf",
"bytes": 69989,
"url": null
},
{
"path": "english/mtl_tokenizer.json",
"sha256": "e7f9364e2c279b2de19f417a83624d9887532a56daec2ddddac470cc71693253",
"bytes": 68125,
"url": null
},
{
"path": "english/s3gen.pt",
"sha256": "9b9ff07e60b20c136e2b1b3d7563a24604e8d2c4c267888d1ee929dd0151d2a3",
"bytes": 1057165844,
"url": null
},
{
"path": "english/s3gen.safetensors",
"sha256": "2b78103c654207393955e4900aac14a12de8ef25f4b09424f1ef91941f161d4e",
"bytes": 1056484620,
"url": null
},
{
"path": "english/s3gen_v3.pt",
"sha256": "f7abce4b196dae2d08d9296cbebc6521b046079577643b42a19a03499d08721e",
"bytes": 1056903694,
"url": null
},
{
"path": "english/s3gen_v3.safetensors",
"sha256": "4a46190f3dccc2230fbb3488a930bccc925862ee68f2662433dfcfe93ce6c2cb",
"bytes": 1056381804,
"url": null
},
{
"path": "english/t3_23lang.safetensors",
"sha256": "5b2194350053f9fc01544e614d923f1e5c8cd785ba596acc142d3d1d8614908c",
"bytes": 2143154168,
"url": null
},
{
"path": "english/t3_cfg.pt",
"sha256": "0b2dd5439fe7e94f379561419847a45bf2c79d0e8ea751c6bbe947ce337789cc",
"bytes": 1064892246,
"url": null
},
{
"path": "english/t3_cfg.safetensors",
"sha256": "914cb1696f47527fe8852ca8f1fe1fa63cb34f76f9c715e84e067b744dd0da81",
"bytes": 2129653744,
"url": null
},
{
"path": "english/t3_mtl23ls_v2.safetensors",
"sha256": "b1237586127ce98e7800a68e49938eb5092846862aabcb6e17b2fda7889a6c75",
"bytes": 2143989752,
"url": null
},
{
"path": "english/t3_mtl23ls_v3.safetensors",
"sha256": "5abca8321ede76f8e61f1cc0d19aea6c946b28871017ce8726f8a69203f05953",
"bytes": 2143989928,
"url": null
},
{
"path": "english/tokenizer.json",
"sha256": "d71e3a44eabb1784df9a68e9f95b251ecbf1a7af6a9f50835856b2ca9d8c14a5",
"bytes": 25470,
"url": null
},
{
"path": "english/ve.pt",
"sha256": "4b16d836bc598509860f6fa068165a8bb5e9ac84f05582dfcf278a5a372879f1",
"bytes": 5698626,
"url": null
},
{
"path": "english/ve.safetensors",
"sha256": "f0921cab452fa278bc25cd23ffd59d36f816d7dc5181dd1bef9751a7fb61f63c",
"bytes": 5695784,
"url": null
},
{
"path": "official-23lang/.gitattributes",
"sha256": "11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361",
"bytes": 1519,
"url": null
},
{
"path": "official-23lang/Cangjie5_TC.json",
"sha256": "7073fd9de919443ae88e0bd2449917a65fe54898a4413ed1edcc4b67f28bce8c",
"bytes": 1920163,
"url": null
},
{
"path": "official-23lang/README.md",
"sha256": "6ac86d5ec7e85a9cf6566a34c6aae60ac1d50a6fceccade4c2aa57cfa6f590b4",
"bytes": 10081,
"url": null
},
{
"path": "official-23lang/conds.pt",
"sha256": "6552d70568833628ba019c6b03459e77fe71ca197d5c560cef9411bee9d87f4e",
"bytes": 107374,
"url": null
},
{
"path": "official-23lang/grapheme_mtl_merged_expanded_v1.json",
"sha256": "69632f47220a788a52ce2661d096453c5655e9bf25289d89a8d832c46ee07dbf",
"bytes": 69989,
"url": null
},
{
"path": "official-23lang/mtl_tokenizer.json",
"sha256": "e7f9364e2c279b2de19f417a83624d9887532a56daec2ddddac470cc71693253",
"bytes": 68125,
"url": null
},
{
"path": "official-23lang/s3gen.pt",
"sha256": "9b9ff07e60b20c136e2b1b3d7563a24604e8d2c4c267888d1ee929dd0151d2a3",
"bytes": 1057165844,
"url": null
},
{
"path": "official-23lang/s3gen.safetensors",
"sha256": "2b78103c654207393955e4900aac14a12de8ef25f4b09424f1ef91941f161d4e",
"bytes": 1056484620,
"url": null
},
{
"path": "official-23lang/s3gen_v3.pt",
"sha256": "f7abce4b196dae2d08d9296cbebc6521b046079577643b42a19a03499d08721e",
"bytes": 1056903694,
"url": null
},
{
"path": "official-23lang/s3gen_v3.safetensors",
"sha256": "4a46190f3dccc2230fbb3488a930bccc925862ee68f2662433dfcfe93ce6c2cb",
"bytes": 1056381804,
"url": null
},
{
"path": "official-23lang/t3_23lang.safetensors",
"sha256": "5b2194350053f9fc01544e614d923f1e5c8cd785ba596acc142d3d1d8614908c",
"bytes": 2143154168,
"url": null
},
{
"path": "official-23lang/t3_cfg.pt",
"sha256": "0b2dd5439fe7e94f379561419847a45bf2c79d0e8ea751c6bbe947ce337789cc",
"bytes": 1064892246,
"url": null
},
{
"path": "official-23lang/t3_cfg.safetensors",
"sha256": "914cb1696f47527fe8852ca8f1fe1fa63cb34f76f9c715e84e067b744dd0da81",
"bytes": 2129653744,
"url": null
},
{
"path": "official-23lang/t3_mtl23ls_v2.safetensors",
"sha256": "b1237586127ce98e7800a68e49938eb5092846862aabcb6e17b2fda7889a6c75",
"bytes": 2143989752,
"url": null
},
{
"path": "official-23lang/t3_mtl23ls_v3.safetensors",
"sha256": "5abca8321ede76f8e61f1cc0d19aea6c946b28871017ce8726f8a69203f05953",
"bytes": 2143989928,
"url": null
},
{
"path": "official-23lang/tokenizer.json",
"sha256": "d71e3a44eabb1784df9a68e9f95b251ecbf1a7af6a9f50835856b2ca9d8c14a5",
"bytes": 25470,
"url": null
},
{
"path": "official-23lang/ve.pt",
"sha256": "4b16d836bc598509860f6fa068165a8bb5e9ac84f05582dfcf278a5a372879f1",
"bytes": 5698626,
"url": null
},
{
"path": "official-23lang/ve.safetensors",
"sha256": "f0921cab452fa278bc25cd23ffd59d36f816d7dc5181dd1bef9751a7fb61f63c",
"bytes": 5695784,
"url": null
}
]
},
"cosyvoice": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/CosyVoice",
"repoPath": "engines/cosyvoice",
"junction": null,
"gate": null,
"origin": {
"repo": "FunAudioLLM/Fun-CosyVoice3-0.5B-2512",
"url": "https://huggingface.co/FunAudioLLM/Fun-CosyVoice3-0.5B-2512"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF API cardData FunAudioLLM/Fun-CosyVoice3-0.5B-2512, fetched 2026-08-18"
},
"postInstall": [
"cosyvoice3-clone-v1 manifest — HARD 30s ceiling on the reference clip, seed-deterministic"
],
"notes": "Task 6 (2026-08-18): Apache-2.0 confirmed via HF API cardData — the manifest's own 'licence not verified from the repo' note is now stale text inside that manifest file (out of scope to edit; this row is the authority). Graduated to mirror: localPath is the PARENT dir, holding exactly the one subfolder Fun-CosyVoice3-0.5B the manifest's own MODELS note names (~9.1GB, ~9.1GB verified non-empty on disk under the shared models root) — the voxcpm/longcat parent-dir pattern.",
"files": [
{
"path": "Fun-CosyVoice3-0.5B/.gitattributes",
"sha256": "42828f8bf9e62763376589790ddf6eccc6bde9b761acd2700dc4a383ade92c33",
"bytes": 1574,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/CosyVoice-BlankEN/config.json",
"sha256": "168aa1bd401abc3bc262ba15ba4e499627a8b4e006e9d050b47c22de20660185",
"bytes": 659,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/CosyVoice-BlankEN/generation_config.json",
"sha256": "e558847a8b4402616f1273797b015104dc266fe4b520056fca88823ba8f8ebe6",
"bytes": 242,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/CosyVoice-BlankEN/merges.txt",
"sha256": "ac8ff86a72bee70828fbc1119bc4398c6f3a9a6e490d7b0dbe917be025478bd0",
"bytes": 1402109,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/CosyVoice-BlankEN/model.safetensors",
"sha256": "130282af0dfa9fe5840737cc49a0d339d06075f83c5a315c3372c9a0740d0b96",
"bytes": 988097824,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/CosyVoice-BlankEN/tokenizer_config.json",
"sha256": "482bd979881423375ca5414e4e0d94cd7c5349dbb17fffd46b4d36d71e62a1bc",
"bytes": 1287,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/CosyVoice-BlankEN/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/README.md",
"sha256": "d4ec0fe1342b60a424ce1cd8971c670d26be38068a96148bcb6a1b8a334c3088",
"bytes": 11982,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/asset/dingding.png",
"sha256": "7f04815e2e676d31b089af6fa270135f3214f2193d5e0ad98b491d007d48f1c6",
"bytes": 122824,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/campplus.onnx",
"sha256": "a6ac6a63997761ae2997373e2ee1c47040854b4b759ea41ec48e4e42df0f4d73",
"bytes": 28303423,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/config.json",
"sha256": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"bytes": 2,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/configuration.json",
"sha256": "c502b6328c67638b401df8dd05de89e9e8d1cff9cd0ada10dfbdbe13556c20de",
"bytes": 47,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/cosyvoice3.yaml",
"sha256": "f5a6b2c6f05139d0f18861a1fe506f751e787026b77c05f7e8fef9f8a4405965",
"bytes": 6934,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/flow.decoder.estimator.fp32.onnx",
"sha256": "9b51b9533a55937762b262bf2cf9c6220ce40760f76d6532cb16a6a6d84059a8",
"bytes": 1326216933,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/flow.pt",
"sha256": "a6fab32a7825e5b0bc855ddd948f8db9370b0a786fbc249caa4595e95b608e4b",
"bytes": 1329116148,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/hift.pt",
"sha256": "b279d7641eb97ae55b3b540cfba4f953c26492a2df758328a89a4d007ab87a65",
"bytes": 83202622,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/llm.pt",
"sha256": "69f43bd545131c30e98947fb360ea8b4dc9916d8e83dded7757c7ea4f5a24970",
"bytes": 2024669519,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/llm.rl.pt",
"sha256": "74d34b01a80c7154670ae75ac372d1b1712c78bceae9f467eb9f1f6f61ec764f",
"bytes": 2024682701,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/speech_tokenizer_v3.batch.onnx",
"sha256": "b156b8a7bbff436585e153f4637b9a368009005ac66efa108a6c8bfb34e5ee43",
"bytes": 969451579,
"url": null
},
{
"path": "Fun-CosyVoice3-0.5B/speech_tokenizer_v3.onnx",
"sha256": "23236a74175dbdda47afc66dbadd5bcb41303c467a57c261cb8539ad9db9208d",
"bytes": 969451503,
"url": null
}
]
},
"dots": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/dots_tts",
"repoPath": "engines/dots",
"junction": null,
"gate": null,
"origin": {
"repo": "dots-studio/dots.tts-base (+ dots.tts-mf, dots.tts-soar)",
"url": "https://huggingface.co/dots-studio/dots.tts-base"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF API cardData, all three repos (dots-studio/dots.tts-base, -mf, -soar), fetched 2026-08-18"
},
"postInstall": [
"dots-tts-clone-v1 manifest — all three variants (~4.9GB each) moved into the shared models root"
],
"notes": "Task 6 (2026-08-18): Apache-2.0 confirmed via HF API cardData on all THREE cited repos. ORG RENAME FOUND: the manifest's origin.repo says 'rednote-hilab' but the live HF API resolves that name to 'dots-studio' (the 'id' field in the JSON response reads dots-studio/dots.tts-base) — HF redirects the old org slug; the origin.repo above is corrected to the current id, origin.url likewise. The manifest's own 'licence not verified from the repo' note is now stale text inside that manifest file (out of scope to edit; this row is the authority). Graduated to mirror: localPath is the PARENT dir holding all three variant subfolders (dots.tts-base, dots.tts-mf, dots.tts-soar) exactly as the manifest's own MODELS note describes (~15.5GB total), verified non-empty on disk under the shared models root. WAVE D (v25.103.0, 2026-08-26): audio.cpp GGUF twins exist (dots-clone-cpp + the dots-tts-cpp text-only roller; DotTTS SOAR q8_0 from audio-cpp/audio.cpp-gguf, sha-locked wave 12). Same Apache-2.0 weights in a GGUF container; NOT staged in OUR mirror - the audio-cpp repo is already their ungated home. The MF and Edit variant packages are not fetched.",
"files": [
{
"path": "dots.tts-base/.gitattributes",
"sha256": "34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930",
"bytes": 1570,
"url": null
},
{
"path": "dots.tts-base/README.md",
"sha256": "09f5e33700dda7e6d56907b740bd0719fe79b6edf78ed295c29449ca6ebe84c9",
"bytes": 6962,
"url": null
},
{
"path": "dots.tts-base/added_tokens.json",
"sha256": "1268ec3933b675b8ff4ddf688f2a6088795a122da8693645bf00a1232c4c8422",
"bytes": 831,
"url": null
},
{
"path": "dots.tts-base/chat_template.jinja",
"sha256": "44d5f08f3f72b837eaad09f13a54c1f9f4eb58d75240334548b7fd52a5437fa5",
"bytes": 2427,
"url": null
},
{
"path": "dots.tts-base/config.json",
"sha256": "d38bd9cff56c9744e33dc0ba62ca39b1ae6ab64561628f8de8132e5b565e9fa5",
"bytes": 1933,
"url": null
},
{
"path": "dots.tts-base/latent_stats.pt",
"sha256": "313b13af56d659ecf869d5f854508fcf823c8f957aefc6bc05244991abd6ffe1",
"bytes": 3197,
"url": null
},
{
"path": "dots.tts-base/llm_config.json",
"sha256": "ce378446b2a6353547ecb44652701ec8ef42610cdcd33f728719033a9d7bb8a1",
"bytes": 2727,
"url": null
},
{
"path": "dots.tts-base/merges.txt",
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
"bytes": 1671853,
"url": null
},
{
"path": "dots.tts-base/model.safetensors",
"sha256": "657ea236f36d03ac86ecc82c0df1aeb368a17c5a5a7ac7369b223ff12776c8bd",
"bytes": 4396289197,
"url": null
},
{
"path": "dots.tts-base/speaker_encoder.safetensors",
"sha256": "1cf3861c9dee79e4db34bd0b8a4155e68bed27a7c6274e168bb6ee4fed191c85",
"bytes": 29150484,
"url": null
},
{
"path": "dots.tts-base/special_tokens_map.json",
"sha256": "52aa5522bb1bce46f653935a856be7d0e006d1acbc881d608d57a8cb687339d7",
"bytes": 1398,
"url": null
},
{
"path": "dots.tts-base/tokenizer.json",
"sha256": "c16521f66774c7a4774e5303b7c8ec5c99830c0be5aef6c6edde3ca2a5e05dd0",
"bytes": 11423263,
"url": null
},
{
"path": "dots.tts-base/tokenizer_config.json",
"sha256": "5fec72f1ad92d8493770ebe6fbd8bf62645f0587bfa2b6e3e12d58c766b92096",
"bytes": 5886,
"url": null
},
{
"path": "dots.tts-base/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "dots.tts-base/vocoder.safetensors",
"sha256": "c0e45c08f480df67ac4c354b465355fcc7e2f6c8765263b6dfeddd1f4671c93d",
"bytes": 723585584,
"url": null
},
{
"path": "dots.tts-mf/.gitattributes",
"sha256": "34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930",
"bytes": 1570,
"url": null
},
{
"path": "dots.tts-mf/README.md",
"sha256": "628ab9d9769c84195b27e000af2cfaea97aa2bb7492f08ad5bc6bcfca4b5b110",
"bytes": 7984,
"url": null
},
{
"path": "dots.tts-mf/added_tokens.json",
"sha256": "1268ec3933b675b8ff4ddf688f2a6088795a122da8693645bf00a1232c4c8422",
"bytes": 831,
"url": null
},
{
"path": "dots.tts-mf/chat_template.jinja",
"sha256": "44d5f08f3f72b837eaad09f13a54c1f9f4eb58d75240334548b7fd52a5437fa5",
"bytes": 2427,
"url": null
},
{
"path": "dots.tts-mf/config.json",
"sha256": "1d6f164cfec8676528776016de4457e0186eceb759a34e9192ee84098a028082",
"bytes": 2091,
"url": null
},
{
"path": "dots.tts-mf/latent_stats.pt",
"sha256": "313b13af56d659ecf869d5f854508fcf823c8f957aefc6bc05244991abd6ffe1",
"bytes": 3197,
"url": null
},
{
"path": "dots.tts-mf/llm_config.json",
"sha256": "ce378446b2a6353547ecb44652701ec8ef42610cdcd33f728719033a9d7bb8a1",
"bytes": 2727,
"url": null
},
{
"path": "dots.tts-mf/merges.txt",
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
"bytes": 1671853,
"url": null
},
{
"path": "dots.tts-mf/model.safetensors",
"sha256": "a16d5798da197bf647fc01915236873e4672e975b0341360703ec49d002c4696",
"bytes": 4398915254,
"url": null
},
{
"path": "dots.tts-mf/speaker_encoder.safetensors",
"sha256": "1cf3861c9dee79e4db34bd0b8a4155e68bed27a7c6274e168bb6ee4fed191c85",
"bytes": 29150484,
"url": null
},
{
"path": "dots.tts-mf/special_tokens_map.json",
"sha256": "52aa5522bb1bce46f653935a856be7d0e006d1acbc881d608d57a8cb687339d7",
"bytes": 1398,
"url": null
},
{
"path": "dots.tts-mf/tokenizer.json",
"sha256": "c16521f66774c7a4774e5303b7c8ec5c99830c0be5aef6c6edde3ca2a5e05dd0",
"bytes": 11423263,
"url": null
},
{
"path": "dots.tts-mf/tokenizer_config.json",
"sha256": "5fec72f1ad92d8493770ebe6fbd8bf62645f0587bfa2b6e3e12d58c766b92096",
"bytes": 5886,
"url": null
},
{
"path": "dots.tts-mf/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "dots.tts-mf/vocoder.safetensors",
"sha256": "c0e45c08f480df67ac4c354b465355fcc7e2f6c8765263b6dfeddd1f4671c93d",
"bytes": 723585584,
"url": null
},
{
"path": "dots.tts-soar/.gitattributes",
"sha256": "34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930",
"bytes": 1570,
"url": null
},
{
"path": "dots.tts-soar/README.md",
"sha256": "3f908474184378d8350d5fbfb8e0669f2019c610b71c4ba6891caf3d079a1701",
"bytes": 8160,
"url": null
},
{
"path": "dots.tts-soar/added_tokens.json",
"sha256": "1268ec3933b675b8ff4ddf688f2a6088795a122da8693645bf00a1232c4c8422",
"bytes": 831,
"url": null
},
{
"path": "dots.tts-soar/chat_template.jinja",
"sha256": "44d5f08f3f72b837eaad09f13a54c1f9f4eb58d75240334548b7fd52a5437fa5",
"bytes": 2427,
"url": null
},
{
"path": "dots.tts-soar/config.json",
"sha256": "d38bd9cff56c9744e33dc0ba62ca39b1ae6ab64561628f8de8132e5b565e9fa5",
"bytes": 1933,
"url": null
},
{
"path": "dots.tts-soar/latent_stats.pt",
"sha256": "313b13af56d659ecf869d5f854508fcf823c8f957aefc6bc05244991abd6ffe1",
"bytes": 3197,
"url": null
},
{
"path": "dots.tts-soar/llm_config.json",
"sha256": "ce378446b2a6353547ecb44652701ec8ef42610cdcd33f728719033a9d7bb8a1",
"bytes": 2727,
"url": null
},
{
"path": "dots.tts-soar/merges.txt",
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
"bytes": 1671853,
"url": null
},
{
"path": "dots.tts-soar/model.safetensors",
"sha256": "2787e6d4fe0b27ac33d28072abcadef53802c841bff037f3ee1f6bb5e1d3a2ce",
"bytes": 4396289197,
"url": null
},
{
"path": "dots.tts-soar/speaker_encoder.safetensors",
"sha256": "1cf3861c9dee79e4db34bd0b8a4155e68bed27a7c6274e168bb6ee4fed191c85",
"bytes": 29150484,
"url": null
},
{
"path": "dots.tts-soar/special_tokens_map.json",
"sha256": "52aa5522bb1bce46f653935a856be7d0e006d1acbc881d608d57a8cb687339d7",
"bytes": 1398,
"url": null
},
{
"path": "dots.tts-soar/tokenizer.json",
"sha256": "c16521f66774c7a4774e5303b7c8ec5c99830c0be5aef6c6edde3ca2a5e05dd0",
"bytes": 11423263,
"url": null
},
{
"path": "dots.tts-soar/tokenizer_config.json",
"sha256": "5fec72f1ad92d8493770ebe6fbd8bf62645f0587bfa2b6e3e12d58c766b92096",
"bytes": 5886,
"url": null
},
{
"path": "dots.tts-soar/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "dots.tts-soar/vocoder.safetensors",
"sha256": "c0e45c08f480df67ac4c354b465355fcc7e2f6c8765263b6dfeddd1f4671c93d",
"bytes": 723585584,
"url": null
}
]
},
"f5tts": {
"source": "origin",
"dest": "hub",
"repoPath": "engines/f5tts",
"junction": null,
"gate": null,
"origin": {
"repo": "SWivid/F5-TTS",
"url": "https://huggingface.co/SWivid/F5-TTS"
},
"licence": {
"weights": "CC-BY-NC-4.0",
"verified": "HF API cardData SWivid/F5-TTS, fetched 2026-08-18"
},
"postInstall": [
"f5tts-clone-v1 manifest — genuinely conditions on the reference transcript (conditions_on_transcript: true)"
],
"notes": "Task 6 (2026-08-18): resolved, NOT graduated — CC-BY-NC-4.0 is non-commercial, fails the graduation bar (permissive-for-redistribution only). Stays origin by design; files exist locally (models\\TTS\\F5-TTS on both trees) but redistribution is the thing the licence forbids, not file availability.",
"files": []
},
"fish2": {
"source": "origin",
"dest": "hub",
"repoPath": "engines/fish2",
"junction": null,
"gate": null,
"origin": {
"repo": "fishaudio/s2-pro (+ drbaph/s2-pro-fp8, jordand/fish-s1-dac-min)",
"url": "https://huggingface.co/fishaudio/s2-pro"
},
"licence": {
"weights": "fish-audio-research-license (custom, non-commercial-only; dependency fish-s1-dac-min is CC-BY-NC-SA-4.0)",
"verified": "HF API cardData all three repos, fetched 2026-08-18"
},
"postInstall": [
"fish2-clone-v1 manifest — five local environment fixes required, see CLAUDE.md 2026-07-21 provisioning addendum"
],
"notes": "Task 6 (2026-08-18): resolved, NOT graduated. fishaudio/s2-pro and drbaph/s2-pro-fp8 both carry license:\"other\", license_name \"fish-audio-research-license\", extra_gated_prompt+extra_gated_fields requiring a 'non-commercial use ONLY' checkbox agreement (top-level 'gated' JSON field itself reads false, but the extra_gated_* block is a real HF consent-gate shape — a human must tick the agreement on the HF page). The third repo, jordand/fish-s1-dac-min (the audio autoencoder dependency), is separately CC-BY-NC-SA-4.0. Both are non-commercial — fails the graduation bar regardless of file availability. WAVE B (v25.101.0, 2026-08-26): audio.cpp GGUF twins exist (fish2-clone-cpp + the fish2-tts-cpp text-only roller; S2 Pro q8_0 from audio-cpp/audio.cpp-gguf, sha-locked wave 10). The SAME research-licensed weights repackaged - the licence above carries; LOCAL use only, never mirrored. NB the cpp route needs NONE of the five comfy env fixes.",
"files": []
},
"higgs-v2": {
"source": "origin",
"dest": "app-models",
"repoPath": "engines/higgs-v2",
"junction": null,
"gate": null,
"origin": {
"repo": "bosonai/higgs-audio-v2-generation-3B-base + bosonai/higgs-audio-v2-tokenizer",
"url": "https://huggingface.co/bosonai/higgs-audio-v2-generation-3B-base"
},
"licence": {
"weights": "BOSON HIGGS AUDIO 2 COMMUNITY LICENSE AGREEMENT (custom, Llama-3-derivative, commercial use above 100k MAU requires a separate agreement)",
"verified": "raw LICENSE file at huggingface.co/bosonai/higgs-audio-v2-generation-3B-base/raw/main/LICENSE, fetched 2026-08-18 (cardData license field itself just says \"other\")"
},
"postInstall": [
"higgs-v2-clone-v1 manifest — SECOND workflow on the 'higgs' kind (v24.94 workflow-match-first fix); needs two model-side fixes, see CLAUDE.md engine environment maintenance"
],
"notes": "Task 6 (2026-08-18): resolved, NOT graduated. Same license family shape as higgs-v3 (see that row) — a Meta-Llama-3-license-derived community agreement, conditional and not a redistribution-permissive licence. Fails the graduation bar regardless of file availability.",
"files": []
},
"indextts2": {
"source": "origin",
"dest": "hub",
"repoPath": "engines/indextts2",
"junction": null,
"gate": null,
"origin": {
"repo": "IndexTeam/IndexTTS-2",
"url": "https://huggingface.co/IndexTeam/IndexTTS-2"
},
"licence": {
"weights": "bilibili Model Use License Agreement (custom, conditional — revenue/user thresholds require a separate licence, high-risk-deployment prohibitions, indemnification terms)",
"verified": "GitHub raw LICENSE at raw.githubusercontent.com/index-tts/index-tts/main/LICENSE, fetched 2026-08-18 (the HF model card itself carries NO license field or file — only the linked GitHub repo, whose own api.github.com license probe reports SPDX NOASSERTION, has the actual terms)"
},
"postInstall": [
"indextts2-clone-v1 + indextts2-emotion-v1 manifests — ref clips TRUNCATE at 15s regardless of adapter advice"
],
"notes": "Task 6 (2026-08-18): resolved, NOT graduated. Custom bilibili licence, not a standard permissive one — fails the graduation bar regardless of file availability. WAVE B (v25.101.0, 2026-08-26): an audio.cpp GGUF twin exists (indextts2-clone-cpp; IndexTTS2 v2 ORIG GGUF from audio-cpp/audio.cpp-gguf, sha-locked wave 10). The SAME bilibili-licensed weights repackaged - the licence above carries; LOCAL use only, never mirrored.",
"files": []
},
"moss": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/moss_tts",
"repoPath": "engines/moss",
"junction": null,
"gate": null,
"origin": {
"repo": "OpenMOSS-Team/MOSS-TTS + OpenMOSS-Team/MOSS-Audio-Tokenizer + OpenMOSS-Team/MOSS-TTS-Local-Transformer + OpenMOSS-Team/MOSS-TTSD-v1.0",
"url": "https://huggingface.co/OpenMOSS-Team/MOSS-TTS"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF API cardData, all four repos (MOSS-TTS, MOSS-Audio-Tokenizer, MOSS-TTS-Local-Transformer, MOSS-TTSD-v1.0), fetched 2026-08-18"
},
"postInstall": [
"moss-tts-clone-v1 manifest — the FULL-SIZE 'v1 8B' delay architecture, distinct from mossnano (100M ONNX, separate row)"
],
"notes": "Task 6 (2026-08-18): Apache-2.0 confirmed on all FOUR OpenMOSS-Team repos actually present on disk (the manifest's origin only named two; the two extras — MOSS-TTS-Local-Transformer and MOSS-TTSD-v1.0 — are the same-author repos the manifest's own notes describe as the 'local' and 'ttsd' architecture variants, and both checked out Apache-2.0 too, so including them is licence-safe). The manifest's own 'licence not verified from the repo' note is now stale text inside that manifest file (out of scope to edit; this row is the authority). Graduated to mirror: localPath is the PARENT dir holding all four subfolders (~47GB total per the manifest's own MODELS note), verified non-empty on disk under the shared models root. Do not conflate with the mossnano row — different model family entirely, same authors.",
"files": [
{
"path": "MOSS-Audio-Tokenizer/.gitattributes",
"sha256": "3153dde53f1819ad009aa95c7a7b7efa9aec86b952fbb5dd79e4903c8f07ee25",
"bytes": 1826,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/README.md",
"sha256": "795c97ec27b38994d4bee033dc3f6f0ad56763510deec3e04a5c1de385139223",
"bytes": 10778,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/__init__.py",
"sha256": "ed0b4910a5e53b1dfc3234cfcf96e2c6890d339f9134a3dfed8481d01b3768fd",
"bytes": 52,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/config.json",
"sha256": "0f669e288d39c9c0ffae4e39babe5167b57e89d3132f0785655d1096a8da8e45",
"bytes": 6601,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/configuration_moss_audio_tokenizer.py",
"sha256": "349b7ff7e1b3f160f9c80df9a0311672b326b8b73e90459122fb39e6878962bf",
"bytes": 12933,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/images/arch.png",
"sha256": "6a7108429fc230ead608a22837a27615aae3458f86e5e47d12fd3bd5d95c7058",
"bytes": 214969,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/images/pesq-nb.png",
"sha256": "107c2f8fb91247264497b50f0bb42f34390ce4593eed28ce4dfea47b88a36797",
"bytes": 743304,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/images/pesq-wb.png",
"sha256": "ae7b264b13e8570c292d8843ce83fcf21e96360e2900c8e19a8ff837be3c90fd",
"bytes": 511823,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/images/sim.png",
"sha256": "79c6340a32e1229ab89fdc71447aa0addf91f146d250b1c3416707c5aacee75d",
"bytes": 492914,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/images/stoi.png",
"sha256": "86d2a427d213fbd8913052d0d16e87c2384a6643f26458314c8771b95eab89c4",
"bytes": 440051,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/model-00001-of-00002.safetensors",
"sha256": "037f441ed30a0ab59f6049de83b824a1b3bd6feb7dbd46c3fbca41fc2f649f28",
"bytes": 4998259168,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/model-00002-of-00002.safetensors",
"sha256": "a187d73d2cda1c2d0676586d9d03c09c0a5813450266af32029c871493fc9582",
"bytes": 2100202560,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/model.safetensors.index.json",
"sha256": "e107e83fee64adc3ccdb993975290cecb00fcf7d72cdaa388011abad16bcc82d",
"bytes": 148113,
"url": null
},
{
"path": "MOSS-Audio-Tokenizer/modeling_moss_audio_tokenizer.py",
"sha256": "65cae7744845f1b8ac65957e918cea508efe331a38e87b882b7530b6c8d7caa5",
"bytes": 70906,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/.gitattributes",
"sha256": "c96474debc0193018d12b76c498fd78b07bc1ebe5dcc31722f75e82f41d76f11",
"bytes": 1634,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/README.md",
"sha256": "60b4fc81cea4310b312450dae6b1fc4700ce51dd477c187512678e17ef28184c",
"bytes": 29532,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/__init__.py",
"sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
"bytes": 0,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/added_tokens.json",
"sha256": "9dfb40d3a3f8654c8af17771d767a494e973b9a33d5c3aeffa1b21f77db8e6d3",
"bytes": 704,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/chat_template.jinja",
"sha256": "894e3ace884e212ae34e1f46063686e7ebcc8b2fa0153af4579dfa5685f63184",
"bytes": 352,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/config.json",
"sha256": "f1d95405f5a37f95a66197daf4a9f425e45eda4cf0a07784ade54ee40a4700ea",
"bytes": 2212,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/configuration_moss_tts.py",
"sha256": "81e6d5125a9ae1c877e348c4339a25480327f4210502cb204e2eb6140cfdef5b",
"bytes": 5997,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/generation_config.json",
"sha256": "a52b5c482438cc48a8bd0ded9b282f6bb5990be8071e944e9cd57e60921b8550",
"bytes": 121,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/inference_utils.py",
"sha256": "12cdf3260114e44c8cba816782f743779ab621e0ba4047e59a3f3579cfcb0e26",
"bytes": 5092,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/merges.txt",
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
"bytes": 1671853,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/model-00001-of-00002.safetensors",
"sha256": "60ec23752c0063ecfafb148d1417d99b7b43c067967652dc90bb9bcb9ea3e243",
"bytes": 4999026432,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/model-00002-of-00002.safetensors",
"sha256": "5045294a3d26c5bbec1dd6a3a6146dc9520cfea35618c5bb80e80a871d8829ad",
"bytes": 1122255376,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/model.safetensors.index.json",
"sha256": "474440fe18124f57b35a63d05a70c0a00029ffd186d7d882b0cba3dd36e09138",
"bytes": 50966,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/modeling_moss_tts.py",
"sha256": "33e9b2ba8f69eab05f3fc096f4d4a61f7a48787dfba2997ba5f55b38cfed1df3",
"bytes": 38194,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/processing_moss_tts.py",
"sha256": "ccb9d7def3d274a27d6c44a975e9c1150298cb69710733512d70d568b630049a",
"bytes": 34730,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/processor_config.json",
"sha256": "6821d4805f10fb8b5cc743c021215097e03f1a25f7c220dfc4fb0d62b5dc97e0",
"bytes": 145,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/special_tokens_map.json",
"sha256": "55cbd0665670e196ce105db3f0d787e5eb35bce235fcb2bd50acbdd60c1e39c6",
"bytes": 631,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/tokenizer.json",
"sha256": "cb3c8fa82993d515469c2800cc455bff4aaa3c4fed9da1f2b0c0668c304f335a",
"bytes": 11422691,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/tokenizer_config.json",
"sha256": "87b1f37a4a73e4c4b61637a1446763e8bdea274347f0418cd5c68ead488c8ea4",
"bytes": 5501,
"url": null
},
{
"path": "MOSS-TTS-Local-Transformer/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "MOSS-TTS/.gitattributes",
"sha256": "c78bc6b9ce15b46697e5daf0f73b888fa1eab303ef8250ad2f9ce686609db5d4",
"bytes": 1612,
"url": null
},
{
"path": "MOSS-TTS/README.md",
"sha256": "4dc69c220b76f1de2a014879f82ee17e0f994d606ce0cbd50a3590feee4ef182",
"bytes": 34624,
"url": null
},
{
"path": "MOSS-TTS/__init__.py",
"sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
"bytes": 0,
"url": null
},
{
"path": "MOSS-TTS/added_tokens.json",
"sha256": "9dfb40d3a3f8654c8af17771d767a494e973b9a33d5c3aeffa1b21f77db8e6d3",
"bytes": 704,
"url": null
},
{
"path": "MOSS-TTS/chat_template.jinja",
"sha256": "894e3ace884e212ae34e1f46063686e7ebcc8b2fa0153af4579dfa5685f63184",
"bytes": 352,
"url": null
},
{
"path": "MOSS-TTS/config.json",
"sha256": "214fc997d98f51ab57925a5939afc6280e76044198b664221622e70d098ed06e",
"bytes": 2275,
"url": null
},
{
"path": "MOSS-TTS/configuration_moss_tts.py",
"sha256": "257842f2fa90124307408f36d55b74e135ee32b7afacbe30280964f7feb329d1",
"bytes": 5584,
"url": null
},
{
"path": "MOSS-TTS/inference_utils.py",
"sha256": "12cdf3260114e44c8cba816782f743779ab621e0ba4047e59a3f3579cfcb0e26",
"bytes": 5092,
"url": null
},
{
"path": "MOSS-TTS/merges.txt",
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
"bytes": 1671853,
"url": null
},
{
"path": "MOSS-TTS/model-00001-of-00004.safetensors",
"sha256": "c1e3c63517c0b6bf1260100c3ea05277ef52b66a8d10ebc6afa50419d27e2e05",
"bytes": 4932667368,
"url": null
},
{
"path": "MOSS-TTS/model-00002-of-00004.safetensors",
"sha256": "7c987704ed6fe458ddffdb1cf00a23319a0c059571f9b6f124c908497d49a4f6",
"bytes": 4915961640,
"url": null
},
{
"path": "MOSS-TTS/model-00003-of-00004.safetensors",
"sha256": "cfd156bc10bac4a38beeb2c559ac4f420d7ee85ae772a29acbac32479a5ab7fc",
"bytes": 4983069760,
"url": null
},
{
"path": "MOSS-TTS/model-00004-of-00004.safetensors",
"sha256": "a591fd0e24a6476a800f2ab1a1cb7b503a0b1e91e322bb81e7b282e7ab13be8e",
"bytes": 2148040304,
"url": null
},
{
"path": "MOSS-TTS/model.safetensors.index.json",
"sha256": "021ef3f74a92ee33fc076dde096d4fa9b7b45c34c9bdbfe730bf79368ac86bb2",
"bytes": 40416,
"url": null
},
{
"path": "MOSS-TTS/modeling_moss_tts.py",
"sha256": "360ceee27c67831d820ff5324f8687f4408db1a91aef53683934544879aa90ca",
"bytes": 25330,
"url": null
},
{
"path": "MOSS-TTS/processing_moss_tts.py",
"sha256": "db33fdd0a1d8176515a93602296dcfeda93ea8f05548659d471db944a8c2f2c1",
"bytes": 34493,
"url": null
},
{
"path": "MOSS-TTS/processor_config.json",
"sha256": "6821d4805f10fb8b5cc743c021215097e03f1a25f7c220dfc4fb0d62b5dc97e0",
"bytes": 145,
"url": null
},
{
"path": "MOSS-TTS/special_tokens_map.json",
"sha256": "55cbd0665670e196ce105db3f0d787e5eb35bce235fcb2bd50acbdd60c1e39c6",
"bytes": 631,
"url": null
},
{
"path": "MOSS-TTS/tokenizer.json",
"sha256": "cb3c8fa82993d515469c2800cc455bff4aaa3c4fed9da1f2b0c0668c304f335a",
"bytes": 11422691,
"url": null
},
{
"path": "MOSS-TTS/tokenizer_config.json",
"sha256": "87b1f37a4a73e4c4b61637a1446763e8bdea274347f0418cd5c68ead488c8ea4",
"bytes": 5501,
"url": null
},
{
"path": "MOSS-TTS/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/.gitattributes",
"sha256": "34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930",
"bytes": 1570,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/README.md",
"sha256": "a8abd4fe44e83e27a4e24d52e6609719c4e83c2465d6a242d339be98a957209f",
"bytes": 20667,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/added_tokens.json",
"sha256": "9dfb40d3a3f8654c8af17771d767a494e973b9a33d5c3aeffa1b21f77db8e6d3",
"bytes": 704,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/chat_template.jinja",
"sha256": "894e3ace884e212ae34e1f46063686e7ebcc8b2fa0153af4579dfa5685f63184",
"bytes": 352,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/config.json",
"sha256": "6e5eebd8d1e3fbc171c2f25ad6568223f1bc5fe0ed9f8c4e3d110c0da5be8e96",
"bytes": 2275,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/configuration_moss_tts.py",
"sha256": "56e31e8572164888ab52e3843a3b0eca8b37355d71bf89ea0d92a4595c1ead0c",
"bytes": 5584,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/generation_config.json",
"sha256": "04e44319786ae9eba99bef56b024543dc2e53a5cd030095fb4ec15412c10ecf8",
"bytes": 132,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/inference_utils.py",
"sha256": "12cdf3260114e44c8cba816782f743779ab621e0ba4047e59a3f3579cfcb0e26",
"bytes": 5092,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/merges.txt",
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
"bytes": 1671853,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/model-00001-of-00004.safetensors",
"sha256": "ad8bb115b6c87c902c76e7a4ef90a6eee98041a87f741fcd163fbe81d855d87a",
"bytes": 4932667368,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/model-00002-of-00004.safetensors",
"sha256": "7cb9641e2f25651a43bc7f681337d79ab57ca50d63355660787712fcee393c40",
"bytes": 4915961640,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/model-00003-of-00004.safetensors",
"sha256": "646b345809967a41308d95afcf5af233ae94de7997befac287017af602c59687",
"bytes": 4983069760,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/model-00004-of-00004.safetensors",
"sha256": "ab0a0ed173b7a86a8cb16b82ca981ad0f1866c2c2d0ecb47c97e72b68ccbaba6",
"bytes": 1879339648,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/model.safetensors.index.json",
"sha256": "9a7229d966d323f16b0142cf9ebdc370923b7a2c5823a73eafef6ed3e1dbcc43",
"bytes": 38448,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/modeling_moss_tts.py",
"sha256": "02b00b7b715f83b5ff658dc8d78fadab9b507edeaa952efdb23c2b998b6c4a4e",
"bytes": 28563,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/processing_moss_tts.py",
"sha256": "24fb7a5495cb92cfd979d15f81fd2554f26a36b4b9f1d089f0436f95c2da14f3",
"bytes": 35797,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/processor_config.json",
"sha256": "6821d4805f10fb8b5cc743c021215097e03f1a25f7c220dfc4fb0d62b5dc97e0",
"bytes": 145,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/special_tokens_map.json",
"sha256": "55cbd0665670e196ce105db3f0d787e5eb35bce235fcb2bd50acbdd60c1e39c6",
"bytes": 631,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/tokenizer.json",
"sha256": "cb3c8fa82993d515469c2800cc455bff4aaa3c4fed9da1f2b0c0668c304f335a",
"bytes": 11422691,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/tokenizer_config.json",
"sha256": "87b1f37a4a73e4c4b61637a1446763e8bdea274347f0418cd5c68ead488c8ea4",
"bytes": 5501,
"url": null
},
{
"path": "MOSS-TTSD-v1.0/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
}
]
},
"omnivoice": {
"source": "origin",
"dest": "hub",
"repoPath": "engines/omnivoice",
"junction": null,
"gate": null,
"origin": {
"repo": "k2-fsa/OmniVoice",
"url": "https://huggingface.co/k2-fsa/OmniVoice"
},
"licence": {
"weights": "CC-BY-NC (code is separately Apache-2.0, but the pre-trained weights are not)",
"verified": "k2-fsa/OmniVoice model card prose, fetched 2026-08-18: 'Our code is released under the Apache 2.0 License. The pre-trained model is licensed under the CC-BY-NC due to constraints from its training data (e.g., Emilia).' — no cardData.license field exists on the HF repo, this is prose-only"
},
"postInstall": [
"omnivoice-clone-v1 + omnivoice-design-v1 manifests — needs pip install --no-deps omnivoice or it renders exactly 1s of silence"
],
"notes": "Task 6 (2026-08-18): resolved, NOT graduated. Weights are CC-BY-NC (the licence a distribution mirror cares about) even though the code is Apache-2.0 — the weights bucket governs here. Fails the graduation bar regardless of file availability. WAVE C (v25.102.0, 2026-08-26): three audio.cpp GGUF twins exist (omnivoice-clone-cpp / omnivoice-tts-cpp / omnivoice-design-cpp; ONE q8_0 file from audio-cpp/audio.cpp-gguf, sha-locked wave 11). The SAME CC-BY-NC weights repackaged - the licence above carries; LOCAL use only, never mirrored. NB the audio-cpp card's row says 'Apache-2.0' and is a MISLABEL - this row's verified reading stands.",
"files": []
},
"qwen3": {
"source": "mirror",
"dest": "comfyui",
"repoPath": "engines/qwen3",
"junction": null,
"gate": null,
"origin": {
"repo": "Qwen/Qwen3-TTS-12Hz-1.7B-Base (+ -CustomVoice, -VoiceDesign, -0.6B variants, Qwen3-TTS-Tokenizer-12Hz)",
"url": "https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-Base"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF API cardData, all five present-on-disk repos (1.7B-Base, 1.7B-CustomVoice, 1.7B-VoiceDesign, 0.6B-Base, 0.6B-CustomVoice) plus Qwen3-TTS-Tokenizer-12Hz, fetched 2026-08-18"
},
"postInstall": [
"qwen3-clone-v1 + qwen3-preset-v1 + qwen3-design-v1 manifests — one kind, three workflows, isolated runtime bootstrap"
],
"notes": "Task 6 (2026-08-18): Apache-2.0 confirmed on every checked repo. Graduated to mirror: array-of-parts localPath covers the five TTS-variant folders actually present on disk (each carries its own embedded speech_tokenizer subfolder, so Qwen3-TTS-Tokenizer-12Hz's content ships redundantly per-part rather than as a sixth part — that mirrors how the pack's own downloader laid them out). DELIBERATELY EXCLUDED: the sibling 'asr' folder (Qwen3-ASR-1.7B, Qwen3-ForcedAligner-0.6B) that sits beside these on disk — those are speech-recognition/alignment utility models, not cited by any qwen3 TTS manifest's origin field, and their licences were not checked this pass; a 0.6B-VoiceDesign variant is also not on disk (not required — nothing here claims full coverage of every possible Qwen3-TTS checkpoint, only what is locally verified and manifest-cited). WAVE B (v25.101.0, 2026-08-26): three audio.cpp GGUF twins exist (qwen3-clone-cpp / qwen3-preset-cpp / qwen3-design-cpp) fetching from audio-cpp/audio.cpp-gguf (Base orig + CustomVoice/VoiceDesign bf16, sha-locked in the adapter's fetch/files.json wave 10). Same Apache-2.0 weights in GGUF containers; they are NOT staged in OUR mirror - the audio-cpp repo is already their ungated home.",
"destPath": {
"Qwen3-TTS-12Hz-1.7B-Base": "TTS/qwen3_tts/Qwen3-TTS-12Hz-1.7B-Base",
"Qwen3-TTS-12Hz-1.7B-CustomVoice": "TTS/qwen3_tts/Qwen3-TTS-12Hz-1.7B-CustomVoice",
"Qwen3-TTS-12Hz-1.7B-VoiceDesign": "TTS/qwen3_tts/Qwen3-TTS-12Hz-1.7B-VoiceDesign",
"Qwen3-TTS-12Hz-0.6B-Base": "TTS/qwen3_tts/Qwen3-TTS-12Hz-0.6B-Base",
"Qwen3-TTS-12Hz-0.6B-CustomVoice": "TTS/qwen3_tts/Qwen3-TTS-12Hz-0.6B-CustomVoice"
},
"files": [
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/.gitattributes",
"sha256": "11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361",
"bytes": 1519,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/README.md",
"sha256": "621b7e88f1867bc03d817176fc1f7f55f4b5a70654b2a07fdcda1e169efb024b",
"bytes": 57817,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/config.json",
"sha256": "b4f01752d15a488abde3e1ab44723ae4f4b9e68a4037257b098b3737893cc1f9",
"bytes": 4494,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/generation_config.json",
"sha256": "f1b90b4513f3b34c62851049e2492d7b4c5940daf1276f89c82b8ef04127f3aa",
"bytes": 245,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/merges.txt",
"sha256": "599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3",
"bytes": 1671839,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/model.safetensors",
"sha256": "38fc7fc51c5e776e840414b6fd443962e9411b9654888fd7913e4da643cb857c",
"bytes": 3857413744,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/preprocessor_config.json",
"sha256": "efdde1022ea9d76928bf7a9cd53139138f5ba2e466e837f08f6105ab1af1c119",
"bytes": 127,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/speech_tokenizer/config.json",
"sha256": "ee65bb901c876664ab8707c487157aa1a6ee57c65969b28fb5ec9dc211e68167",
"bytes": 2336,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/speech_tokenizer/configuration.json",
"sha256": "6bc26d64eb5024b4d1dab5a52371958b429256d6c9d59787f1f5294a54e0cebd",
"bytes": 76,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/speech_tokenizer/model.safetensors",
"sha256": "836b7b357f5ea43e889936a3709af68dfe3751881acefe4ecf0dbd30ba571258",
"bytes": 682293092,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/speech_tokenizer/preprocessor_config.json",
"sha256": "fcb3805e597e786d4067706e602f6688524640f8d3396790e2e09b5942fcbdfb",
"bytes": 234,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/tokenizer_config.json",
"sha256": "dc3c31c3bdaedd5016382bb3cbe07323026775ad51f5a4fb564505992ae4a670",
"bytes": 7344,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-Base/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/.gitattributes",
"sha256": "11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361",
"bytes": 1519,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/README.md",
"sha256": "64f65e809b51cc0c35f393fbbfcc2d735c0cb3fdbbc6f3fdc4a5e6ce55e9d088",
"bytes": 57846,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/config.json",
"sha256": "17a07f527a1c25ea30b4e023a184482a23d3e279d697b1dc81b1bde498d29cf9",
"bytes": 4908,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/generation_config.json",
"sha256": "f1b90b4513f3b34c62851049e2492d7b4c5940daf1276f89c82b8ef04127f3aa",
"bytes": 245,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/merges.txt",
"sha256": "599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3",
"bytes": 1671839,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/model.safetensors",
"sha256": "38b1d5971bdbd982b561cccec982669a53b0537c3cf5e9bd4778ed07bb2f5137",
"bytes": 3833402552,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/preprocessor_config.json",
"sha256": "efdde1022ea9d76928bf7a9cd53139138f5ba2e466e837f08f6105ab1af1c119",
"bytes": 127,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/speech_tokenizer/config.json",
"sha256": "ee65bb901c876664ab8707c487157aa1a6ee57c65969b28fb5ec9dc211e68167",
"bytes": 2336,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/speech_tokenizer/configuration.json",
"sha256": "6bc26d64eb5024b4d1dab5a52371958b429256d6c9d59787f1f5294a54e0cebd",
"bytes": 76,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/speech_tokenizer/model.safetensors",
"sha256": "836b7b357f5ea43e889936a3709af68dfe3751881acefe4ecf0dbd30ba571258",
"bytes": 682293092,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/speech_tokenizer/preprocessor_config.json",
"sha256": "fcb3805e597e786d4067706e602f6688524640f8d3396790e2e09b5942fcbdfb",
"bytes": 234,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/tokenizer_config.json",
"sha256": "dc3c31c3bdaedd5016382bb3cbe07323026775ad51f5a4fb564505992ae4a670",
"bytes": 7344,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-CustomVoice/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/.gitattributes",
"sha256": "11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361",
"bytes": 1519,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/README.md",
"sha256": "acfaf6c0d433866cb5e2bb73e7915ea3767f5a2e3abca413b485635a2c72b5e6",
"bytes": 3214,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/config.json",
"sha256": "aecd2cc4c1fe9edef1cb7ca7c401685a43879ad43f3f9e883f1c6760b61731e0",
"bytes": 4421,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/generation_config.json",
"sha256": "f1b90b4513f3b34c62851049e2492d7b4c5940daf1276f89c82b8ef04127f3aa",
"bytes": 245,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/merges.txt",
"sha256": "599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3",
"bytes": 1671839,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/model.safetensors",
"sha256": "391e8db219f292c515297cdceeb43e4eae67cdde35fa57e79a6a8a532fca0522",
"bytes": 3833402552,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/preprocessor_config.json",
"sha256": "efdde1022ea9d76928bf7a9cd53139138f5ba2e466e837f08f6105ab1af1c119",
"bytes": 127,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/speech_tokenizer/config.json",
"sha256": "ee65bb901c876664ab8707c487157aa1a6ee57c65969b28fb5ec9dc211e68167",
"bytes": 2336,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/speech_tokenizer/configuration.json",
"sha256": "6bc26d64eb5024b4d1dab5a52371958b429256d6c9d59787f1f5294a54e0cebd",
"bytes": 76,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/speech_tokenizer/model.safetensors",
"sha256": "836b7b357f5ea43e889936a3709af68dfe3751881acefe4ecf0dbd30ba571258",
"bytes": 682293092,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/speech_tokenizer/preprocessor_config.json",
"sha256": "fcb3805e597e786d4067706e602f6688524640f8d3396790e2e09b5942fcbdfb",
"bytes": 234,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/tokenizer_config.json",
"sha256": "dc3c31c3bdaedd5016382bb3cbe07323026775ad51f5a4fb564505992ae4a670",
"bytes": 7344,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-1.7B-VoiceDesign/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/.gitattributes",
"sha256": "11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361",
"bytes": 1519,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/README.md",
"sha256": "181187b6057906bd960bc7f938d0b7a16652509776a0d52c4885b4ae5ccda0ea",
"bytes": 3640,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/config.json",
"sha256": "2e714c787c8edb98b05432685cddb634add2de4d4e645f653d68251ef72ba011",
"bytes": 4494,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/generation_config.json",
"sha256": "f1b90b4513f3b34c62851049e2492d7b4c5940daf1276f89c82b8ef04127f3aa",
"bytes": 245,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/merges.txt",
"sha256": "599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3",
"bytes": 1671839,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/model.safetensors",
"sha256": "180b3b10eb1c9f1b4db7806d5475bae3071c0243c299d49926bab1da3b6946f6",
"bytes": 1829344272,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/preprocessor_config.json",
"sha256": "efdde1022ea9d76928bf7a9cd53139138f5ba2e466e837f08f6105ab1af1c119",
"bytes": 127,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/speech_tokenizer/config.json",
"sha256": "ee65bb901c876664ab8707c487157aa1a6ee57c65969b28fb5ec9dc211e68167",
"bytes": 2336,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/speech_tokenizer/configuration.json",
"sha256": "6bc26d64eb5024b4d1dab5a52371958b429256d6c9d59787f1f5294a54e0cebd",
"bytes": 76,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/speech_tokenizer/model.safetensors",
"sha256": "836b7b357f5ea43e889936a3709af68dfe3751881acefe4ecf0dbd30ba571258",
"bytes": 682293092,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/speech_tokenizer/preprocessor_config.json",
"sha256": "fcb3805e597e786d4067706e602f6688524640f8d3396790e2e09b5942fcbdfb",
"bytes": 234,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/tokenizer_config.json",
"sha256": "dc3c31c3bdaedd5016382bb3cbe07323026775ad51f5a4fb564505992ae4a670",
"bytes": 7344,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-Base/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/.gitattributes",
"sha256": "11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361",
"bytes": 1519,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/README.md",
"sha256": "1e3adeecc7a72d6756fdb77c2847f8e994195e105b51206a7a5c049b0dfa48a8",
"bytes": 3263,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/config.json",
"sha256": "81aca2b6fac304944d8acf345272d8a9a727d5fc2e2e66b222ab4729340c7455",
"bytes": 4908,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/generation_config.json",
"sha256": "f1b90b4513f3b34c62851049e2492d7b4c5940daf1276f89c82b8ef04127f3aa",
"bytes": 245,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/merges.txt",
"sha256": "599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3",
"bytes": 1671839,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/model.safetensors",
"sha256": "bc3c7e785eb961179c25450d1acff03f839e0002f2f3a5aeb67b5735c0fa2adb",
"bytes": 1811626576,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/preprocessor_config.json",
"sha256": "efdde1022ea9d76928bf7a9cd53139138f5ba2e466e837f08f6105ab1af1c119",
"bytes": 127,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/speech_tokenizer/config.json",
"sha256": "ee65bb901c876664ab8707c487157aa1a6ee57c65969b28fb5ec9dc211e68167",
"bytes": 2336,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/speech_tokenizer/configuration.json",
"sha256": "6bc26d64eb5024b4d1dab5a52371958b429256d6c9d59787f1f5294a54e0cebd",
"bytes": 76,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/speech_tokenizer/model.safetensors",
"sha256": "836b7b357f5ea43e889936a3709af68dfe3751881acefe4ecf0dbd30ba571258",
"bytes": 682293092,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/speech_tokenizer/preprocessor_config.json",
"sha256": "fcb3805e597e786d4067706e602f6688524640f8d3396790e2e09b5942fcbdfb",
"bytes": 234,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/tokenizer_config.json",
"sha256": "dc3c31c3bdaedd5016382bb3cbe07323026775ad51f5a4fb564505992ae4a670",
"bytes": 7344,
"url": null
},
{
"path": "Qwen3-TTS-12Hz-0.6B-CustomVoice/vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833,
"url": null
}
]
},
"stepaudio": {
"source": "origin",
"dest": "hub",
"repoPath": "engines/stepaudio",
"junction": null,
"gate": null,
"origin": {
"repo": "stepfun-ai/Step-Audio-EditX + stepfun-ai/Step-Audio-Tokenizer",
"url": "https://huggingface.co/stepfun-ai/Step-Audio-EditX"
},
"licence": {
"weights": "UNVERIFIED",
"verified": null
},
"postInstall": [
"step-editx-clone-v1 manifest — needed FunASR-Paraformer + two files copied up from CosyVoice-300M-25Hz; SLOWEST engine in the roster (~100s warm)"
],
"notes": "verification attempt recorded 2026-08-18: the HF repo stepfun-ai/Step-Audio-EditX carries NO cardData.license field and NO LICENSE file in its file list (checked via ?full=true — siblings has no LICENSE entry). The model card's own prose says 'The code in this open-source repository is licensed under the Apache 2.0 License' and links to a GitHub repo; api.github.com/repos/stepfun-ai/Step-Audio-EditX independently confirms Apache-2.0 (SPDX) for that GitHub repo. But that statement and that check are both about the CODE, not the WEIGHTS sitting in the HF repo — the weights repo itself ships no licence of its own, which is a genuine ambiguity, not an unresolved lookup. Recorded as UNVERIFIED rather than assumed Apache-2.0, and NOT graduated on that ambiguity; a future task could ask the owner/upstream directly rather than infer weights-inherit-code-licence.",
"files": []
},
"vibevoice": {
"source": "mirror",
"dest": "comfyui",
"destPath": "TTS/VibeVoice/vibevoice-1.5B",
"repoPath": "engines/vibevoice",
"junction": null,
"gate": null,
"origin": {
"repo": "microsoft/VibeVoice-1.5B (+ aoi-ot/VibeVoice-Large, FranckyB/VibeVoice-Large)",
"url": "https://huggingface.co/microsoft/VibeVoice-1.5B"
},
"licence": {
"weights": "MIT",
"verified": "HF API cardData, all three repos (microsoft/VibeVoice-1.5B, aoi-ot/VibeVoice-Large, FranckyB/VibeVoice-Large), fetched 2026-08-18"
},
"postInstall": [
"vibevoice-clone-v1 manifest — multi_speaker_mode pinned 'Native Multi-Speaker'; 'Custom Character Switching' drops every line after the first"
],
"notes": "Task 6 (2026-08-18): MIT confirmed on all three cited repos. Graduated to mirror, but ONLY the vibevoice-1.5B subfolder — the manifest's own default model_name is 'vibevoice-1.5B' and that is the exact folder matched here, verified non-empty (~5.1GB, config.json + three safetensors shards). Two SIBLING folders exist on disk beside it (VibeVoice/vibevoice-7B and VibeVoice/kugelaudio-0-open) and are DELIBERATELY EXCLUDED — vibevoice-7B is a different, larger checkpoint (aoi-ot's org is presumed to correspond given the pack combo names 'vibevoice-1.5B'/'vibevoice-7B', but that correspondence was not independently confirmed against a specific HF repo id, and kugelaudio-0-open is a THIRD PARTY fine-tune (Kugel) that was not licence-checked at all this pass. Only the exactly-matched, exactly-verified subfolder is mirrored. WAVE C (v25.102.0, 2026-08-26): an audio.cpp GGUF twin exists (vibevoice-clone-cpp; 1.5B bf16 GGUF from audio-cpp/audio.cpp-gguf, sha-locked wave 11). Same MIT weights in a GGUF container; NOT staged in OUR mirror - the audio-cpp repo is already their home. NB the cpp route requires Speaker-N-labelled text.",
"files": [
{
"path": ".gitattributes",
"sha256": "cd0e76e59343ef6743c4b32c45719af8ed1b140ebbd3c199782f839dd4fb3bb1",
"bytes": 1603,
"url": null
},
{
"path": "README.md",
"sha256": "9a881db6567b6091dae09a56fc31edc6647ba04f74c575fa99c838b72c68879d",
"bytes": 7279,
"url": null
},
{
"path": "config.json",
"sha256": "b0d6db52f45dd5179b42217c55112aa2ee66e7ad7b40b0950833c9f6780a0f21",
"bytes": 2762,
"url": null
},
{
"path": "figures/Fig1.png",
"sha256": "64464f28380f81e76c88d76431a08b48c7f82a283e17f2e32c241c4b03407d19",
"bytes": 153971,
"url": null
},
{
"path": "model-00001-of-00003.safetensors",
"sha256": "c5f0a61ddeaeb028e3af540ba4dee7933ad30f9f30b6e1320dd9c875a2daa033",
"bytes": 1975317828,
"url": null
},
{
"path": "model-00002-of-00003.safetensors",
"sha256": "81c3891f7b2493eb48a9eb6f5be0df48d4f1a4bfd952d84e21683ca6d0bf7969",
"bytes": 1983051688,
"url": null
},
{
"path": "model-00003-of-00003.safetensors",
"sha256": "cb6e7e5e86b4a41fffbe1f3aaf445d0d50b5e21ed47574101b777f77d75fa196",
"bytes": 1449832938,
"url": null
},
{
"path": "model.safetensors.index.json",
"sha256": "067db9b10fdecee3a5588aa00206794156c7125f5e85f3f2234e0e6d821ee629",
"bytes": 122616,
"url": null
},
{
"path": "preprocessor_config.json",
"sha256": "56bef7fb56db168e31dff685337c4acb6a35b5ce9c68d255128b01395b442c43",
"bytes": 351,
"url": null
}
]
},
"zonos": {
"source": "origin",
"dest": "hub",
"repoPath": "engines/zonos",
"junction": null,
"gate": null,
"origin": {
"repo": "Zyphra/Zonos-v0.1-hybrid (+ -transformer, -speaker-embedding)",
"url": "https://huggingface.co/Zyphra/Zonos-v0.1-hybrid"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF API cardData, all three repos (Zyphra/Zonos-v0.1-hybrid, -transformer, -speaker-embedding), fetched 2026-08-18"
},
"postInstall": [
"zonos-clone-v1 manifest — the original v1 engine (ComfyUI-Zonos node), distinct from zonos2 (separate mirror row above)"
],
"notes": "Task 6 (2026-08-18): licence resolved permissive (Apache-2.0, same org and same verdict as the already-mirrored zonos2 row) — but NOT graduated this pass. The local copy at models\\zonos\\Zonos-v0.1-hybrid is a live git-lfs CHECKOUT, not a plain snapshot: it carries a .git\\ directory measured at 3.1GB (56 objects) sitting inside the same folder as the actual weights. Pointing localPath at it as-is would sweep 3.1GB of git internals into the mirror lock/upload, which build-lock.js's walker cannot exclude (it hashes every file under localPath). Per the graduation rule ('licence-verified-permissive but files NOT locally stageable → stays origin'): needs a clean re-stage (e.g. a plain hf download or an rsync excluding .git) before this can graduate. Zonos-v0.1-transformer, by contrast, IS a clean snapshot on disk with no .git — only the hybrid variant has this problem. Do not conflate with the zonos2 row — different pack, different model generation.",
"files": []
},
"kitten": {
"source": "mirror",
"dest": "comfyui",
"destPath": "kittentts/KittenML_kitten-tts-mini-0.8/models--KittenML--kitten-tts-mini-0.8/snapshots/c02725660cea441db4c383af69f1f26f5cd00947",
"repoPath": "engines/kitten",
"junction": null,
"gate": null,
"origin": {
"repo": "KittenML/kitten-tts-mini-0.8",
"url": "https://huggingface.co/KittenML/kitten-tts-mini-0.8"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF API cardData KittenML/kitten-tts-mini-0.8, fetched 2026-08-18"
},
"postInstall": [
"kitten-tts-v1 manifest — Saganaki22/ComfyUI-KittenTTS (Apache-2.0 pack), model prefetched by the wheel's own downloader; the CPU-vs-GPU device note in the manifest is STALE per the 2026-08-18 measurement (see CLAUDE.md owner decisions owed)"
],
"notes": "Task 6 (2026-08-18): Apache-2.0 confirmed via HF API cardData. Graduated to mirror: localPath points directly at the wheel's own huggingface_hub snapshot dir (config.json, kitten_tts_mini_v0_8.onnx, voices.npz — three files via NTFS symlinks into the sibling blobs\\ dir, the exact hub-cache shape build-lock.js's walk() classify() function already resolves through statSync). JUNCTION FIELD CORRECTED: the row previously claimed a junction { link, target: null } but on inspection there is no per-engine junction here — ComfyUI\\models is itself a symlink to the shared models root, and 'kittentts\\KittenML_kitten-tts-mini-0.8' is a real subdirectory reached straight through that outer symlink (same shape as the kokoro and mossnano rows, which correctly carry junction: null); the placeholder claim was inaccurate and is fixed rather than carried forward. destPath CORRECTED post-final-review (2026-08-18): it keeps the full hub-cache interior (models--KittenML--.../snapshots/<rev>) because the kitten wheel manages this dir AS a hub cache — files placed flat at the truncated path would not be found and the wheel would re-download (the neutts/voxtream/pocket-preset precedent).",
"files": [
{
"path": "config.json",
"sha256": "6b160bc9b19e24ecb21e84bc14f8a7da21fdf47ec72d42450bc5cf514b61804a",
"bytes": 470,
"url": null
},
{
"path": "kitten_tts_mini_v0_8.onnx",
"sha256": "0f5bbae4fc4800c98dbc544a87ecfa79510de2fb8222db30d12e5bfe9177df91",
"bytes": 78268016,
"url": null
},
{
"path": "voices.npz",
"sha256": "40ad2638952b77b7b2f30127e2608e169fc69dd256b53bd8aaa3409a33193c42",
"bytes": 3278902,
"url": null
}
]
},
"voxtream": {
"source": "mirror",
"dest": "hub",
"destPath": "models--herimor--voxtream2/snapshots/49addec130217e8e9e82a6f49437c315c5c851fc",
"repoPath": "engines/voxtream",
"junction": null,
"gate": null,
"origin": {
"repo": "herimor/voxtream2",
"url": "https://huggingface.co/herimor/voxtream2"
},
"licence": {
"weights": "CC-BY-4.0",
"verified": "HF API cardData herimor/voxtream2, fetched 2026-08-18"
},
"postInstall": [
"voxtream-clone-v1 manifest — the pack's FIRST isolated-runtime engine (its own venv, spawned per render)"
],
"notes": "Task 6 (2026-08-18): CC-BY-4.0 confirmed via HF API cardData — permissive, attribution required (matches the graduation bar the same way pocket-preset's CC-BY-4.0 already does). Graduated to mirror: localPath points at the HF hub cache snapshot dir (config.json, model.safetensors, phoneme_to_token.json — three files via the same hub-cache symlink shape as kitten/neutts/pocket-preset), verified non-empty.",
"files": [
{
"path": "config.json",
"sha256": "e13431ca36e6f2cb213c373eafdec47901bb490b0b9e84db675b608c1b247daa",
"bytes": 418,
"url": null
},
{
"path": "model.safetensors",
"sha256": "b0761a350f9908227dcdce4556328a5896d1bab9d609939869ea941f206febb5",
"bytes": 1851507776,
"url": null
},
{
"path": "phoneme_to_token.json",
"sha256": "10c9cb4e2762c1f0f98d092b23cdef109342c9fbb6b16ae087ae30fd8f7c79f7",
"bytes": 2737,
"url": null
}
]
},
"dia2": {
"source": "mirror",
"dest": "comfyui",
"destPath": "Dia2",
"repoPath": "engines/dia2",
"junction": null,
"gate": null,
"origin": {
"repo": "lord-lethris/ComfyUI-lethris-dia2 (pack) + nari-labs/Dia2-2B (weights, identified via HF search this task — the manifest names only the GitHub pack repo)",
"url": "https://huggingface.co/nari-labs/Dia2-2B"
},
"licence": {
"weights": "Apache-2.0",
"verified": "HF API cardData nari-labs/Dia2-2B, fetched 2026-08-18 — identified by searching HF for author=nari-labs (the manifest itself does not name the weights repo, only the GitHub pack)"
},
"postInstall": [
"dia2-tts-v1 manifest — Nari Labs Dia2-2B dialogue cloner, UNGATED, 7.68GB prefetched into models\\Dia2; pack fixed by a marked fork registering folder key 'dia2'"
],
"notes": "Task 6 (2026-08-18): weights repo identified (nari-labs/Dia2-2B, alongside a smaller nari-labs/Dia2-1B — both Apache-2.0) and licence confirmed Apache-2.0. Graduated to mirror: localPath verified non-empty (model.safetensors 7,678,277,416 bytes — matches the manifest's own '7.68GB prefetched byte-verified' claim almost exactly, plus tokenizer/config files, all flat in one dir, no junction needed per the manifest's own folder-key-registration note).",
"files": [
{
"path": "added_tokens.json",
"sha256": "e6407562f41f7ca0948082446ef6439c0bb2b2c46a314df17f067ecdc0c8c160",
"bytes": 1156,
"url": null
},
{
"path": "config.json",
"sha256": "d59c6899abcbfeb54b954c6bfe95285822a39446456f0a4faabd7dc262de135a",
"bytes": 1439,
"url": null
},
{
"path": "dia2_assets.json",
"sha256": "87416e14135885d83257388b5b02174062ec34ef24edf1c32d18618fa85dc3bc",
"bytes": 125,
"url": null
},
{
"path": "merges.txt",
"sha256": "0b54e8aa4e53d5383e2e4bc635a56b43f9647f7b13832d5d9ecd8f82dac4f510",
"bytes": 466391,
"url": null
},
{
"path": "model.safetensors",
"sha256": "b3eb6b4758beaae97dcfa87d430706ede0d78306cc3afb1eaf4397f216abd1db",
"bytes": 7678277416,
"url": null
},
{
"path": "special_tokens_map.json",
"sha256": "c1fd3b95eeae0a8d7e6f5d4f893d6a657a32eef6a6a6b5d941c742e844df70af",
"bytes": 7923,
"url": null
},
{
"path": "tokenizer.json",
"sha256": "7931199d153b72035a0b3acf7bfd276266248a0b6c5e9a5a25d233ca1424846c",
"bytes": 3532285,
"url": null
},
{
"path": "tokenizer_config.json",
"sha256": "02b56f0084d1192d2402abb2885f509db5cfe93a855fc05816ea5e2d4ff27171",
"bytes": 13337,
"url": null
},
{
"path": "vocab.json",
"sha256": "82b84012e3add4d01d12ba14442026e49b8cbbaead1f79ecf3d919784f82dc79",
"bytes": 800662,
"url": null
}
]
},
"elevenlabs": {
"source": "none",
"dest": null,
"repoPath": "engines/elevenlabs",
"junction": null,
"gate": null,
"origin": null,
"licence": {
"weights": "n/a",
"verified": null
},
"postInstall": [],
"notes": "cloud API — nothing to host. Direct engine descriptor lives in main.js (DIRECT_DESCRIPTORS.elevenlabs), not a workflow folder.",
"files": []
},
"windows": {
"source": "none",
"dest": null,
"repoPath": "engines/windows",
"junction": null,
"gate": null,
"origin": null,
"licence": {
"weights": "n/a",
"verified": null
},
"postInstall": [],
"notes": "OS voices via SAPI5 (parrot-sapi.ps1) — nothing to host. Direct engine descriptor lives in main.js (DIRECT_DESCRIPTORS.windows).",
"files": []
},
"minimax": {
"source": "origin",
"dest": "pack",
"repoPath": "engines/minimax",
"junction": null,
"gate": null,
"origin": {
"repo": "audio-cpp/audio.cpp-gguf (MiniMax-H3-Q4-GGUF; original MiniMaxAI/MiniMax-H3)",
"url": "https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/MiniMax-H3-Q4-GGUF"
},
"licence": {
"weights": "MiniMax-H3 Community License (TERRITORY-RESTRICTED)",
"verified": "the audio-cpp/audio.cpp-gguf card row for MiniMax-H3-Q4-GGUF, fetched 2026-08-26; OWNER RULING 2026-08-26 ('minimax_h3 - yes, install it'): accepted for LOCAL use only"
},
"postInstall": [
"fetched by functions/tts-and-sts-engines/audiocpp-adapter/fetch/fetch.js wave 13 (18 entries, 32.6 GB sha-verified: the default q4_k package only - dit.gguf sha256 f122a03b9e04 prefix + text_encoder_q4_k + the folded audio VAE + video VAE + FL2VA processor sidecars; dit_int8 and the unfolded audio VAE are NOT fetched)"
],
"notes": "WAVE G (v25.106.0, 2026-08-26 - the campaign closer): minimax-h3-cpp, the ONE task-gen route - a text-to-audio-SCENE generator on the 32.4 GB default q4_k package (sha-locked wave 13; the dit_int8 variant and the unfolded audio VAE are NOT fetched). NEVER mirrored - the licence is territory-restricted; local use per the owner's ruling. The family's music/sfx tasks are the Music-tab lane. NB the binary REFUSES --task tts for this family (probed) - gen is the real task.",
"files": []
}
}
}