Unverified50% confidenceBenchmarkExact time
Llama 3.1 405B在MMLU上的得分已接近GPT-4o的水平，Qwen 2.5 72B在数学和代码生成基准上某些指标超越了GPT-4 Turbo
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/7/2026
First Seen
Valid until: 12/6/2026
Sources
Related Entities
Related Claims
UnverifiedQwen3-4B在多项推理类评测指标上已与DeepSeek V3或GPT-4o相差无几76% similarUnverifiedQwen2.5-72B 在多项基准测试中接近 Llama-3.1-405B 的表现72% similarUnverifiedModels like GPT-4 and Claude 3.5 surpassed 90% accuracy on MMLU, significantly diminishing its discriminative power71% similarUnverifiedGlimmer智能指数35分，超越Gemma 2 27B（30分），几乎追平Qwen 2.5（36分），Qwen 2.5 72B以38分居榜首70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/871702API
curl https://kongchang.com/api/v1/knowledge/claims/871702MCP
get_claim(id=871702)