Yesterday
-
catalog8.json: min app 1.6.0
-
catalog8.json: H3 description without the RAM claim
-
catalog8.json: MiniMax-H3-Image (apps 1.5.7+)
3 days ago
-
catalog7: common512 (text encoder for 512 tokens) for apps 1.5.4+
-
common512: text encoder for 512 tokens (apps 1.5.4+); common/: back to the 128 token files for older apps
-
catalog: Qwen-Image 2.1 is a 7B model (was 20B)
-
common (v69): text encoder slices for 512 tokens (were 128)
-
catalog6: MiaoMax RealSkin and Harem MOSS packs
4 days ago
-
catalog5: Qwen description for the base + LoRA scheme
-
catalog5: plain Qwen-Image-2.1-Turbo is one pack (manifestbase.json)
-
catalog4: MOSS-Image 8 step add-on; catalog5.json: packs + models with modes
-
catalog4: Harem 1.6 turbo/base, 3D Harem base
-
Anima 2B pairs: Harem 1.6 turbo (new), 3D Harem LH3D 1.1 base (new), Harem 1.6 base renamed to harem16base
-
catalog4: base models, Qwen 8 step
-
Anima base packs (no turbo): Harem 2.9B/2B, RealSkin 2.9B/2B, int8 blocks, fp16 activations
-
2B models (3D Harem, RealSkin): int8 block weights instead of fp16 (4.2 GB -> 2.6 GB)
6 days ago
-
catalog3: add Mage-Flow-Edit-Turbo (app 1.4.1+)
-
Add Qwen-Image 2.1 Turbo (needs app 1.4.0+)
8 days ago
-
Add catalog2.json (adds MOSS-Image; catalog.json unchanged for older apps)
9 days ago
-
Upload README.md with huggingface_hub
-
Upload folder using huggingface_hub
10 days ago
-
Upload NOTICE.txt with huggingface_hub
-
Upload README.md with huggingface_hub
-
Upload LICENSE.md with huggingface_hub
-
Upload NOTICE.txt with huggingface_hub
-
Upload README.md with huggingface_hub
-
Upload README.md with huggingface_hub
-
Upload catalog.json with huggingface_hub
-
Upload folder using huggingface_hub
-
initial commit
