Unverified50% confidenceBenchmarkExact time
The Q3_XXS model paired with an F16 KV cache scores approximately 80% on the needle test, while Q3_K_XL paired with a kvarn3 KV cache scores close to 100%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/7/2026
First Seen
Valid until: 12/6/2026
Sources
Related Entities
Related Claims
UnverifiedIn the Q4_K_XL scheme, MoE layers use MXFP4 and the remaining layers are compressed to Q8_0, achieving 96% same on top-1% prediction consistency71% similarUnverifiedThe Q4_K_M quantization scheme averages approximately 4.8 bits per weight, while the Q8_0 quantization scheme uses 8 bits per weight.70% similarVerifiedQwen3.5's parameter activation rate is 4.3%, while DeepSeek V3's 671B model has an activation rate of 5.5%.69% similarUnverified4Bit量化下模型的基准测试得分通常能保持原始精度的90%-95%68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/871780API
curl https://kongchang.com/api/v1/knowledge/claims/871780MCP
get_claim(id=871780)