Verified65% confidenceFactExact time
Using the release of GPT-4 as a reference point, closed-source models once led the open-source camp by more than a year.
3
Sources
65%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
Unverified2023年末开源模型在 SWE-bench 上与 GPT-4 差距显著，闭源模型解决率约为开源模型的2-3倍78% similarUnverified2024年初多个声称对标GPT-4的开源模型因训练数据疑似包含顶级闭源模型输出而引发质疑，OpenAI和Anthropic相继加强对异常访问模式的检测76% similarVerifiedGPT-5.5是OpenAI于2025年中期发布的模型，相比GPT-4o在推理深度和上下文理解方面有显著提升75% similarVerifiedTool-calling capabilities were first formally introduced by OpenAI in June 2023 in the GPT-4 API, after which mainstream models including Claude and Gemini successively followed suit75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/592698API
curl https://kongchang.com/api/v1/knowledge/claims/592698MCP
get_claim(id=592698)