已过期50% 置信事实精确时间
Composer 2.5在编程能力基准测试上已基本追平Anthropic和OpenAI的顶级模型，而前代Composer 2与顶级模型差距明显
1
来源数
50%
置信度
中期 (~90 天)
时效性
2026/7/2
首次发现
有效期至：2026/9/30(已过期)
来源
Cursor Composer 2.5 Hands-On: An AI Coding Model That's Faster and 10x Cheaper
bilibili星海研究2026/5/21
相关事实
待验证LLaVA、InternVL、Qwen-VL等开源视觉语言模型在多项基准测试中已接近甚至超越部分闭源模型的表现64% 相似待验证OpenAI在5.4版本确认不再单独训练Codex专属模型，其能力已折叠进主力模型64% 相似待验证OpenAI Codex performed requirements analysis before writing code, offering two technical approaches: a full solution and an MVP approach.62% 相似待验证在AIME 2024和CodeForces等权威基准测试中，DeepSeek R1的表现与OpenAI O1正式版不相上下62% 相似待验证OpenAI offers Codex models with different parameter scales and capability tiers, where lightweight models are suitable for simple code completion and formatting tasks, and high-parameter models perform better in complex logical reasoning, multi-file refactoring, and architecture design.61% 相似
引用此条事实
Stable URI
https://kongchang.com/claim/59461API
curl https://kongchang.com/api/v1/knowledge/claims/59461MCP
get_claim(id=59461)