Qwen-Image-2512 — INT8 tensorwise (+ConvRot), stock-ComfyUI native
Offline INT8 quantization of the Qwen/Qwen-Image-2512 transformer, produced with
Comfy-Org/comfy-quants
(export-model-int8-tensorwise). Loads natively in stock ComfyUI >= v0.27.0
(QUANT_ALGOS["int8_tensorwise"], SM >= 7.5 / Turing+) — no custom node required.
Transformer-only (for models/diffusion_models/, load via UNETLoader); pair with
the usual Qwen-Image text encoder and VAE.
Quantization contract
- 839 quantized Linears (selection mirrors the family's official recipe — same membership as the mxfp8/nvfp4 exports); all 839 ConvRot-rotated (regular Hadamard, group 256).
- Per layer:
int8weight +float32 [out,1]scale +comfy_quantmarker{"format": "int8_tensorwise", "convrot": true, "convrot_groupsize": 256}(byte-exact to stock ComfyUI's save path). - Quant math bit-faithful to comfy-kitchen >= 0.2.15.