Unverified50% confidenceBenchmarkExact time
By the end of 2024, open-source models such as Llama 3.1 405B and Qwen2.5-72B had approached or in some benchmarks even surpassed GPT-4 level performance.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/7/2026
First Seen
Valid until: 12/6/2026
Sources
Related Entities
Related Claims
Unverified在文档摘要基准测试上，Qwen2.5-32B-Instruct和Llama-3.1-70B的量化版本已能达到GPT-4早期版本85%-92%的表现水平78% similarVerifiedMeta's released Llama 3.1 405B has approached GPT-4 levels across multiple benchmark tests.78% 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 3.5.77% similarUnverifiedOpenAI's model naming follows a traceable public release pattern, from GPT-3 (2020) and GPT-3.5 (2022) to GPT-4 (2023) and GPT-4o (2024).77% similarUnverifiedOn mainstream benchmarks such as MMLU, HumanEval, and GSM8K, open-source models have reached 85–95% of the level of early versions of GPT-4.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/873084API
curl https://kongchang.com/api/v1/knowledge/claims/873084MCP
get_claim(id=873084)