Unverified50% confidenceBenchmarkExact time
Open-weight models have already matched or even surpassed the top closed-source models from one year ago on multiple benchmarks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/21/2026
First Seen
Valid until: 12/20/2026
Sources
Related Entities
Related Claims
UnverifiedOpen-source models' performance on benchmarks has increasingly approached, and in some tasks even surpassed, that of closed-source models.78% similarUnverifiedIFEval基准测试专门评估模型遵循可验证指令的能力，测试显示即使最强闭源模型在严格约束条件下的遵循率也难以达到90%以上66% similarUnverifiedOpenAI采用宪法AI思路对输出进行价值对齐，并通过红队测试在发布前挖掘模型危险能力边界66% similarUnverifiedOn standard code benchmarks such as HumanEval and MBPP, top open-source models have already reached 85–95% of the performance level of closed-source models.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/939494API
curl https://kongchang.com/api/v1/knowledge/claims/939494MCP
get_claim(id=939494)