Unverified50% confidenceBenchmarkExact time
Q4_K_M量化文件大小17.77GB，pp2048速度96.8 tok/s，tg128速度11.4 tok/s
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/20/2026
First Seen
Valid until: 11/18/2026
Sources
Related Claims
VerifiedThe Qwen2.5:7B model is approximately 4 GB in size after Q4 quantization.80% similarUnverifiedIQ4_XS量化文件大小15.57GB，pp2048为95.4 tok/s，tg128为10.9 tok/s78% similarUnverifiedOn the RTX 5090, the Qwen3.6 27B model's prefill speed increases from 174 to 253 tokens/s, with a VRAM usage of approximately 18 GB.76% similarUnverified使用IQ4_XS配合-fa 1 -ctk q8_0 -ctv q8_0对KV缓存进行q8_0量化后，可加载完整32k上下文，常驻内存约16.6GB75% similarUnverifiedUD-Q4_K_XL量化版本体积155GB，Top-1精度为96%75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/778776API
curl https://kongchang.com/api/v1/knowledge/claims/778776MCP
get_claim(id=778776)