Expired85% confidenceFactTime unknown
Only 3 models scored above 90 in the BugFind test, with the best performance coming from the APEX-quantized 35B MoE.
1
Sources
85%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedIn the BugFind code bug-finding test, Qwen3.6 27B Q4 precision scored 87 points, higher than Q6's 80 points, exhibiting a quantization reversal phenomenon.63% similarUnverified混元3是混合专家模型（MoE），总参数量为295B，激活参数约21B62% similarUnverifiedQwen3's 235B-parameter MoE flagship model approaches or surpasses closed-source top-tier models on multiple authoritative benchmarks.62% similarUnverifiedMuse-Glimmer-30B在测试中通过了88%的题目61% similarUnverified千问3.6的35B MOE模型每次只激活其中3B的参数61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/17967API
curl https://kongchang.com/api/v1/knowledge/claims/17967MCP
get_claim(id=17967)