OpenAI o3
OpenAI o3 is an AI model published by OpenAI, released on 2025-04-16, for 推理大模型, with 0.0B parameters, and 200K tokens context length, under the 不开源 license.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Modality | Input | Output |
|---|---|---|
| Text | $10 | $40 |
| Image | $10 | -- |
OpenAI o3 currently shows benchmark results led by Creative Writing (2 / 22, score 87.65), Aider-Polyglot (3 / 26, score 81.30), MATH-500 (5 / 43, score 98.10). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.
OpenAI o3是当前OpenAI最先进的推理大模型。作为o系列旗舰模型，该模型在复杂问题解决、跨领域分析和视觉推理任务中树立了新的性能标杆，尤其擅长需要深度逻辑推演的多步骤工作流。
基准测试o1o3(无工具)o3(全工具)
AIME 2025数学竞赛79.2%88.9%98.4%
博士级科学问题(GPQA)8.12%20.32%24.90%
视觉数学推理(MathVista)55.1%78.6%-
代码编辑任务(Aider)64.4%81.3%-
在同等延迟条件下，o3推理深度较o1提升3.2倍，复杂问题解决成功率提高42%。
该模型标志着AI系统向自主工具调度与跨模态推理的重要突破，为复杂决策场景提供新的技术基座。
欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送
