Individual skill
awq-quantization
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
backend/cli/skills/ml-training/awq/SKILL.md
Source commit a08e0ad0ad3d6448123039ee4659163eb2e35d99
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Network or broad shell
- Last checked
- Nov 2, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 4,351
- Repository last updated
- Nov 6, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent Repository topic
- ai Repository topic
- bun Repository topic
- cli Repository topic
- llm Repository topic
- ml Repository topic
- open-source Repository topic
- research Repository topic
- science Repository topic
- ai-agent Repository topic
- optimization Declared skill tag
- awq Declared skill tag
- quantization Declared skill tag
- 4-bit Declared skill tag
- activation-aware Declared skill tag
- memory-optimization Declared skill tag
- fast-inference Declared skill tag
- vllm-integration Declared skill tag
- marlin-kernels Declared skill tag
- research Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesSKILL.mdcore · 8,700 bytesreferences/advanced-usage.mddoc · 7,983 bytesreferences/troubleshooting.mddoc · 7,733 bytes
Read excerpt
# AWQ (Activation-aware Weight Quantization)
4-bit quantization that preserves salient weights based on activation patterns, achieving 3x speedup with minimal accuracy loss.
## When to use AWQ
**Use AWQ when:**
- Need 4-bit quantization with <5% accuracy loss
- Deploying instruction-tuned or chat models (AWQ generalizes better)
- Want ~2.5-3x inference speedup over FP16
- Using vLLM for production serving
- Have Ampere+ GPUs (A100, H100, RTX 40xx) for Marlin kernel support
**Use GPTQ instead when:**