Unverified50% confidenceBenchmarkExact time
GPT-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
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/13/2026
First Seen
Valid until: 10/11/2026
Sources
GPT-5.6发布：三款模型对标Claude，ChatGPT Work超级应用全解析
bilibiliAI-seeker7/9/2026
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82% 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81% similarUnverified据非官方爆料，Terminal Bench 3.0基准测试中Claude Opus 5 Max以43.5%的得分超越GPT-5.6-Sol的34.6%重登榜首80% similarUnverified在Agent's Last Exam基准测试中，GPT-5.6 Soul（extra high档）得分近54%，Fable（max档）为45%79% similarPartially VerifiedClaude Opus 4.8在SWE-bench基准测试中得分69.2%，GPT 5.5为58.6%，Google Gemini为54.2%78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/501552API
curl https://kongchang.com/api/v1/knowledge/claims/501552MCP
get_claim(id=501552)