openai/o3-pro
上下文长度: 200,000
text+image->text
Other
2025-06-11 更新
The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently better answers. Note that BYOK is required for this model. Set up here: https://openrouter.ai/settings/integrations