sm079/anima-studio

Anima Turbo for the browser

Weights for Anima Studio, which runs Anima (Turbo v1.1, by CircleStone Labs and Comfy Org) entirely in a web browser with WebGPU. The page downloads these files once, keeps them in the browser's storage, and generates images on the visitor's own graphics card. Nothing is sent to a server.

Unofficial. These are modified (repackaged and quantized) versions of Anima, not an official CircleStone Labs release. Non-commercial use only: see LICENSE.md and NOTICE.

Files

Part	Original	Compressed (int8)	Extra compressed (w4a8)
Image model (Anima Turbo v1.1, 2B)	4.18 GB	2.42 GB	1.71 GB
Text encoder (Qwen3 0.6B)	1.19 GB	0.61 GB	0.44 GB
Image decoder (Qwen-Image VAE, decoder only)	51 MB

manifest.json lists the files, their sizes and the recommended settings (8 steps, ER-SDE, beta schedule, shift 3); tokenizers/ holds the Qwen and T5 tokenizers.

• int8: ComfyUI int8_tensorwise with ConvRot (256-group Hadamard rotation), as in Comfy-Org's MiniMax-H3 int8_convrot release.
• w4a8: ComfyUI asym_w4a8_int8 ("mixed": first and last image-model blocks stay int8), as in Kijai's MiniMax-H3 w4a8_mixed release.
• The decoder's causal 3D convolutions are collapsed to 2D for single images. The text encoder's final layer stays bf16 in the compressed versions.

License

The CircleStone Model is licensed by CircleStone Labs LLC under the CircleStone Non-Commercial License. Copyright CircleStone Labs LLC. See NOTICE for the full attribution notice, the NVIDIA Open Model License (Anima is a Derivative Model of Cosmos-Predict2-2B-Text2Image), and the Apache 2.0 Qwen components.