Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.
安装
npx skills add davila7/claude-code-templates --skill gptqnpx skills add davila7/claude-code-templates --skill gptq -a claude-code -gnpx skills add davila7/claude-code-templates --skill gptq -a codex -ggemini skills install https://github.com/davila7/claude-code-templates.git --path cli-tool/components/skills/ai-research/optimization-gptqgh skill install davila7/claude-code-templates gptqgit clone --depth 1 https://github.com/davila7/claude-code-templates /tmp/davila7__claude-code-templates
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/davila7__claude-code-templates/cli-tool/components/skills/ai-research/optimization-gptq ~/.claude/skills/gptq # Claude Code
cp -R /tmp/davila7__claude-code-templates/cli-tool/components/skills/ai-research/optimization-gptq ~/.agents/skills/gptq # Codex, Cursor, Gemini CLI, Copilot, OpenCode