Expired50% confidenceFactExact time
At the end of 2023, there was a significant gap in programming capabilities between open-source models and GPT-4; by 2024–2025, domestic open-source models such as DeepSeek Coder, Qwen Coder, and the Kimi K series had substantially narrowed that gap.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/7/2026
First Seen
Valid until: 10/5/2026(expired)
Sources
Kimi K2.7 Code Hands-On: How Does High Speed Mode Feel in Practice?
bilibili御风大世界6/23/2026
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82% similarUnverifiedSince the second half of 2024, the performance of open-source models has closely approached that of closed-source models, with the ELO scores of Llama, DeepSeek, Zhipu GLM, and other open-source models nearing those of GPT-4o and Claude 3.5.76% similarUnverified以Qwen、DeepSeek、Llama为代表的开源/开放权重模型阵营正快速缩小与GPT-4o、Claude等顶级闭源模型的差距76% similarUnverifiedOpen-source models such as Meta's Llama 3.1 series, DeepSeek-V3, and Alibaba's Qwen2.5-Coder have significantly narrowed the gap with GPT-4-level models on code generation benchmarks such as HumanEval and SWE-bench.75% similarUnverified2024年下半年，Meta的Llama系列、DeepSeek、智谱GLM等开源模型的ELO积分已逼近GPT-4o和Claude 3.5的区间，分差压缩至个位数71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/164262API
curl https://kongchang.com/api/v1/knowledge/claims/164262MCP
get_claim(id=164262)