Qwen3.6 35B A3B int4 AutoRound EC Extended Calibration (EC) INT4 AutoRound quantization of Qwen/Qwen3.6 35B A3B, a 35B MoE (3B active, 128 experts) multimodal model. Drop in replacement for Intel/Qwen3.6 35B A3B int4 AutoRound with wider calibration settings for improved quality on long context and reasoning heavy workloads. Calibration — Extended vs Intel default Intel (v0.13.0) EC (this model) iters 200 (default) 400 nsamples 128 (default) 256 seqlen 512 (default) 4096 batch size 8 (default) 8 (default) grad accum 1 (default) 1 (default) ignore layers mtp.fc mtp.fc bits / group 4 / 128 4 / 128 sym true true packing format auto round:auto gptq auto round:auto gptq Intel does not publish their iters/nsamples/seqlen in the released artifact or README. Our EC values are 2× Intel's CLI defaults on iters and nsamples , and 8× on seqlen — targeting better activation coverage for long context and multimodal workloads. Environment Component Version auto round 0.13.0 (merge commit 2dda035 — PR 1705) transformers 5.5.1 (matches Intel's exact pin for this variant) torch 2.6.0+cu124 safetensors 0.7.0 huggingface hub 1.11.0 Hardware RunPod H200 SXM (1× 141 GB HBM3e) Reproducible invocation No…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy