| { | |
| "format": "maya3-safetensors", | |
| "format_version": 1, | |
| "model_type": "maya3", | |
| "model": { | |
| "backbone": "backbone", | |
| "attn": "sdpa", | |
| "speaker_anchor": { | |
| "enabled": true, | |
| "hidden": 512, | |
| "input_dim": 192, | |
| "dropout": 0.2 | |
| }, | |
| "depth_grouped": true, | |
| "depth_ckpt_from": 9 | |
| }, | |
| "key_prefix": "model.", | |
| "aliases": { | |
| "backbone.lm.lm_head.weight": "backbone.lm.model.embed_tokens.weight" | |
| }, | |
| "tensors": { | |
| "count": 732, | |
| "stored": 731, | |
| "dtypes": { | |
| "bfloat16": 732 | |
| }, | |
| "stored_bytes": 5004797312, | |
| "n_params_unique": 2502398656 | |
| }, | |
| "architecture": { | |
| "backbone": "Qwen3-1.7B (Qwen3ForCausalLM, built from backbone/config.json)", | |
| "depth_decoder": "DepthV2, 31 levels", | |
| "speaker_anchor": "ECAPA-192 -> MLP" | |
| }, | |
| "audio": { | |
| "sample_rate": 24000, | |
| "frame_rate_hz": 12.5, | |
| "n_codebooks": 32, | |
| "codebook_size": 2048, | |
| "codec": "kyutai/mimi" | |
| }, | |
| "speaker_encoder": { | |
| "repo": "speechbrain/spkrec-ecapa-voxceleb", | |
| "dim": 192 | |
| }, | |
| "languages": [ | |
| "as", | |
| "bn", | |
| "en", | |
| "gu", | |
| "hi", | |
| "kn", | |
| "ml", | |
| "mr", | |
| "or", | |
| "pa", | |
| "ta", | |
| "te" | |
| ], | |
| "source": { | |
| "name": "maya3-oss-v1", | |
| "update": null, | |
| "format": null, | |
| "sha256_total": null, | |
| "hash_alg": null, | |
| "n_params": null | |
| }, | |
| "converted_utc": "2026-10-26T04:09:25Z", | |
| "converter": { | |
| "torch": "2.8.0+cu128", | |
| "safetensors": "0.8.0" | |
| } | |
| } |