Unverified50% confidenceBenchmarkExact time
PrismML将Qwen 3.6从原始约54GB压缩到不足4GB，压缩比超过13倍
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/14/2026
First Seen
Valid until: 10/12/2026
Sources
Related Claims
UnverifiedPrismo宣称其方案让270亿参数全部同时激活，并将原版54GB的QW3.6模型压缩至不到4GB，压缩比超过13倍且声称不影响性能81% similarUnverified原版QW3.6 27B模型文件体积约为54GB，Prismo通过数学压缩技术压缩至不到4GB，压缩比超过13倍，且声称不影响模型性能77% similarUnverifiedQ3_K_M将每个权重压缩至约3bit，27B模型权重体积可降至约12～13GB73% similarUnverifiedQwen3系列编程模型约20GB加载，推理时可膨胀到35-40GB71% similarVerifiedQwen2.5:7B模型Q4量化后文件大小约为4GB70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/504222API
curl https://kongchang.com/api/v1/knowledge/claims/504222MCP
get_claim(id=504222)