Expired65% confidenceOpinionTime unknown
The latest open-source models have reached or even surpassed GPT-4-level performance on multiple benchmarks for specific tasks.
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedStanford University's AgentBench benchmark test in 2024 showed that GPT-4-level models have an end-to-end success rate below 30% on complex tasks requiring cross-system coordination.78% similarUnverifiedOpenAI's research indicates that carefully designed prompts can improve GPT-4's performance on specific tasks by 30%–50%.78% similarUnverifiedGPT-5、Grok 4等模型在基准测试（如HumanEval、SWE-bench）上表现优异，但这些测试通常是单次代码生成，与需要多轮工具调用的真实工程任务存在显著差距77% similarUnverifiedIn benchmark tests officially published by OpenAI, GPT-5.6 achieved a new top score on Terminal Bench in the programming domain and uses fewer tokens than GPT-5.5 on biology benchmarks (the results were self-disclosed by OpenAI and have not yet been independently verified).77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/33646API
curl https://kongchang.com/api/v1/knowledge/claims/33646MCP
get_claim(id=33646)