Expired50% confidenceBenchmarkExact time
In the second half of 2024, the ELO scores of open-source models such as Meta's Llama series, DeepSeek, and Zhipu GLM had approached the range of GPT-4o and Claude 3.5, with the score gap compressed to single digits.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/22/2026
First Seen
Valid until: 10/20/2026(expired)
Sources
Related Claims
UnverifiedSince 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 3.5.84% similarUnverifiedThe overall capability gap among mainstream large models such as GPT-4o, Claude 3.5, and DeepSeek V3 has narrowed to within 5%.72% similarVerifiedBetween 2023 and 2024, the performance gap among mainstream large language models such as GPT-4, Claude 3, and Gemini Ultra continued to narrow on benchmarks, making them difficult to distinguish on most general tasks.71% similarUnverifiedAt 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.71% similarUnverifiedThe concept of intelligent model routing only became feasible for engineering after 2023, when the capability gap between flagship models such as GPT-4, Claude, and Gemini and open-source small models such as Llama and Mistral was quantified by benchmark tests.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/592667API
curl https://kongchang.com/api/v1/knowledge/claims/592667MCP
get_claim(id=592667)