Unverified50% confidenceBenchmarkExact time
Ornith 9B 在 SWE-Bench 上得分 69.4,接近千问 3.5-35B 的 70 分
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/21/2026
First Seen
Valid until: 10/19/2026
Sources
Related Claims
Unverified千问3.6 Plus综合得分69.0分,排名末位被淘汰出局64% similarUnverified通义千问3.6在SWE-Bench测试中得分78.8分,对比Claude Opus的80.9分64% similarUnverified在SWE-Bench基准测试中,Codex得分在69-80分之间60% similarUnverified在SWE-Bench Multilingual测试中,HY3得分75.8,略微落后于DeepSeek V4 Pro的76.258% similarUnverifiedOrnith 35B MoE 约 30 秒给出三神问题的正确答案,而 Ornith 9B 和原始千问 3.5-9B 均未通过58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/578088API
curl https://kongchang.com/api/v1/knowledge/claims/578088MCP
get_claim(id=578088)