Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.
Install
npx skills add davila7/claude-code-templates --skill quantizing-models-bitsandbytesnpx skills add davila7/claude-code-templates --skill quantizing-models-bitsandbytes -a claude-code -gnpx skills add davila7/claude-code-templates --skill quantizing-models-bitsandbytes -a codex -ggemini skills install https://github.com/davila7/claude-code-templates.git --path cli-tool/components/skills/ai-research/optimization-bitsandbytesgh skill install davila7/claude-code-templates quantizing-models-bitsandbytesgit 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-bitsandbytes ~/.claude/skills/quantizing-models-bitsandbytes # Claude Code
cp -R /tmp/davila7__claude-code-templates/cli-tool/components/skills/ai-research/optimization-bitsandbytes ~/.agents/skills/quantizing-models-bitsandbytes # Codex, Cursor, Gemini CLI, Copilot, OpenCode