Views
No views yet
diff_dec = high-noise expert, diff_dec_low = low-noise expert),
quantized to fp8 e4m3 scaled in the ComfyUI format.Comfy-Org/Bernini-R's
wan2.2_bernini_r_*_fp8_scaled.safetensors (verified: same 1815 keys, shapes,
dtypes, and __metadata__) — the difference is only the weights, which here are
the full Bernini renderer (jointly trained with the MLLM planner) rather than
the renderer-only Bernini-R.| File | model_type | size |
|---|---|---|
wan2.2_bernini_high_noise_fp8_scaled.safetensors | bernini_high | ~15.5 GB |
wan2.2_bernini_low_noise_fp8_scaled.safetensors | bernini_low | ~15.5 GB |
ComfyUI/models/diffusion_models/ and use them anywhere the
Bernini-R fp8_scaled files work (same model_type, same keys).comfy_quant = {"format": "float8_e4m3fn"}.self_attn.{q,k,v,o}, cross_attn.{q,k,v} (cross-attn o
kept in fp16), ffn.0, ffn.2 — 9 per block × 40 = 360 weights per expert.W: scale = max(|W|)/448,
W_fp8 = (W/scale).clamp(±448).to(float8_e4m3fn), stored alongside a scalar
weight_scale (fp32). Dequant: W ≈ W_fp8.to(dtype) * weight_scale.modulation, patch_embedding, text/time_embedding,
time_projection, head, all biases) is kept in fp16.ByteDance/Bernini-Diffusers
(bernini/ checkpoint, fp32), with diffusers WanTransformer3DModel keys
remapped to the original Wan / ComfyUI naming.