Unverified65% confidenceFactExact time
Qwen3 Next Flash的4-bit量化版本（Q4K XL）约111GB
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
9/6/2026
First Seen
Valid until: 12/5/2026
Sources
Related Entities
Related Claims
UnverifiedQwen-7B经过4-bit量化后，一张搭载24GB显存的消费级显卡即可流畅运行76% similarUnverifiedQwen 3.6 27B FP8模型在4×3080Ti 16GB GPU上成功完成本地部署74% similarUnverifiedQwythos-9B的Q4量化版本仅需4GB显存即可本地部署运行74% similarVerifiedQwen2.5:7B模型Q4量化后文件大小约为4GB73% similarUnverifiedDeepSeek V4 Flash的UD-Q8_K_XL量化版本文件体积为162GB，采用MXFP4+BF16混合精度，达到完全无损73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/863365API
curl https://kongchang.com/api/v1/knowledge/claims/863365MCP
get_claim(id=863365)