Stabhappy/kroma-v0.2-base-INT8-convrot

Quantized derivative of kroma-v0.2-base.safetensors. Not an official Krea/Kroma release. Original weights © their respective owners under the Krea 2 Community License.

File

Quantization method

• Format: INT8 ConvRot — row-wise INT8 weights with group-wise Hadamard rotation (group size 256) applied to smooth activation outliers before quantization. Targets near-lossless quality (≈ GGUF-Q8 class).
• Optimization: SVD learned rounding + bias correction (calibration samples: 3072, tool default).
• Output: native ComfyUI INT8 — comfy_quant metadata embedded per-layer. Loads with the standard model loader, no custom node required.
• Inference compute: INT8 tensor cores (torch._int_mm) + Triton kernels.

Command

Layer strategy (Krea 2 profile, --krea2)

• INT8-quantized: the 28 main DiT blocks (blocks.0–27) — attention (wq/wk/wv/wo/gate) and MLP (gate/up/down).
• Kept high-precision (F32): first, last (incl. norm/linear/modulation), tmlp, tproj, txtfusion (layerwise + refiner blocks), txtmlp — the outlier-sensitive embedding / time / text-fusion layers.

Requirements

• ComfyUI with native INT8 support (Comfy-Org commit 1a510f04 or recent release).
• NVIDIA GPU with sufficient INT8 TOPS — RTX 30-series and up.
• Text encoder: qwen3vl_4b_fp8_scaled.safetensors → models/text_encoders/
• VAE: qwen_image_vae.safetensors (Qwen VAE) → models/vae/
• Triton installed in the ComfyUI environment.

Usage

Notes

• This file is the DiT transformer only; text encoder and VAE load separately as above.
• LoRA: works via the standard loader. For best LoRA quality, merge the LoRA into the base before quantizing (pre-lora) rather than stacking onto the INT8 weights.
• ConvRot = a QuaRot-derived plug-and-play rotation; small inference overhead in exchange for a quality gain over plain row-wise INT8.
• Retains the original model's license and attribution requirements; disclose AI-generated outputs where required.

Credits