quark-onnx-autosearch-pro

SkillSearch

End-to-end Quark ONNX AutoSearchPro recipe — drives `quark.onnx.AutoSearchPro` (Optuna-based hyperparameter search) on a `.onnx` model to find the best quantization config (activation/weight spec, calibration method, CLE, AdaRound / AdaQuant, FastFinetune params). Use when the user wants to "auto search", "tune quantization", "find the best quant config", "sweep AdaRound/AdaQuant", "run AutoSearchPro / AutoSearch", "two-stage search", or pick one of the built-in presets (`ADVANCED_SEARCH`, `XINT8_SEARCH`, `A8W8_SEARCH`, `A16W8_SEARCH`) for their `.onnx` model. Not for HuggingFace / safetensors / PyTorch input models — use quark-torch-ptq instead. Not for single-shot ONNX PTQ without search — use quark-onnx-ptq.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the quark-onnx-autosearch-pro skill

What this skill tells your AI

The instructions your AI receives, as published by amd/quark in .claude/skills/quark-onnx-autosearch-pro/SKILL.md and read by ahel’s review.

Read and follow the instructions in .claude/skills-impl/l3-recipes/onnx/quark-onnx-autosearch-pro/SKILL.md.

Signals

GitHub stars
166
Forks
33
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
quark-onnx-autosearch-pro
Source
github.com/amd/quark