Unverified50% confidenceFactExact time
AWQ由MIT团队提出，发现仅约1%的权重对激活值影响显著，通过保护这部分关键权重实现更优的低比特量化效果
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
生产级LLM部署指南：模型选型、GPU配置与推理引擎全解析
Related Claims
UnverifiedThe core insight of AWQ is that only approximately 1% of salient weights have a major impact on model output, and protecting them enables model quality to be maintained at extremely low bit widths84% similarUnverifiedAWQ（Activation-aware Weight Quantization）于2023年提出，发现权重中约 1% 的显著通道对精度影响最大69% similarUnverifiedLoRA微调只需训练不到1%的参数即可达到接近全参微调的效果，显存需求大幅降低至单张消费级显卡可承受范围63% similarVerifiedWhen fine-tuning with LoRA, the actual number of parameters that need to be trained is only 0.1%-1% of the original.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/503790API
curl https://kongchang.com/api/v1/knowledge/claims/503790MCP
get_claim(id=503790)