Unverified60% confidenceBenchmarkExact time
GPT-5.6 Sol在Humanity's Last Exam取得最高分，且价格远低于Claude Opus和Fable
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/11/2026
First Seen
Valid until: 10/9/2026
Sources
GPT-5.6全解析：Sol旗舰模型性能、短板与竞争力评测
bilibili小牛AI_XNAI7/10/2026
Related Claims
Unverified在Agents' Last Exam评测中，GPT-5.6 Sol取得53.6分，比采用自适应推理的Claude Fable 5高出13.1分74% similarUnverifiedGPT-5.6在Agent's Last Exam测试中以53.6分创下历史新高，领先Claude Opus 4.5达13.1分74% similarUnverified在长周期智能体工作流测试（Engines Last Exam）中，GPT-5.6 Sol 得分 53.6，比 Faber-5 高出 13.1 分72% similarUnverifiedOpus 4.7每次解题的中位输出token数为60,000，而GPT 5.5仅需16,00071% similarUnverifiedOpus 4.7在长上下文（50万-100万token）检索方面表现明显退步，远不如DeepSeek和GPT 5.571% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/486152API
curl https://kongchang.com/api/v1/knowledge/claims/486152MCP
get_claim(id=486152)