Unverified50% confidenceBenchmarkExact time
GPT-3.5-Turbo、GPT-4、Claude等主流模型在长上下文检索任务中的性能均呈现U形分布
1
Sources
50%
Confidence
Long-term
Relevance
7/17/2026
First Seen
Sources
Related Claims
UnverifiedThe GPT-4 series or Claude has the most mature tool-calling capabilities, with high stability in complex instruction parsing.83% similarUnverifiedGPT-4 and Claude 3.5 are cited as examples of sufficiently capable underlying models.83% similarPartially VerifiedAll three GPT-5.6 models achieved a perfect score of 10 on the long-horizon agent fine-tuning task, while GPT-5.5 and Opus 4.7 could only score two or three points on that task.83% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/538955API
curl https://kongchang.com/api/v1/knowledge/claims/538955MCP
get_claim(id=538955)