Unverified50% confidenceBenchmarkExact time
在AI Builder模型能力智力榜上，Claude Opus 5.5 Max以58分登顶，小米MiMo V2.6 Pro以46分列第9，Qwen 3.8 Max以45分列第10
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/30/2026
First Seen
Valid until: 10/14/2026
Sources
Related Entities
Related Claims
UnverifiedArtificial Analysis 给 Qwen 3.8 打出52的智能指数，追平满推理档的 GPT-5.6，智能体分51，高于 Claude Opus 4.874% similarUnverified在SWE Bench Pro等智能体编程基准中，Qwen 3.8据称在部分项目上超越了Opus 4.6 Max72% similarUnverified在Artificial Analysis智能指数中Grok 4.7（超高思考强度）排第五位得分46，前四名为Fable 5.1、Astra、Cloud Opus 5和MuseSpark 1.3 Max70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/960783API
curl https://kongchang.com/api/v1/knowledge/claims/960783MCP
get_claim(id=960783)