From flagship reasoning to full multimodal — pick the right model for every workload.
Flagship reasoning model with thinking mode. Best for complex analysis, multi-step reasoning, and advanced coding tasks.
Balanced multimodal model supporting text, image, and video. Ideal for production workloads.
Fast and cost-effective with 1M context. Great for high-throughput pipelines.
Ultra-fast and ultra-cheap. Perfect for real-time chat and high-volume inference.
Specialized for code generation, debugging, and technical documentation.
Advanced vision-language model. Understands images, charts, documents, and screenshots.
Lighter vision model for cost-sensitive visual tasks.
Full multimodal: text, image, and audio input. Native speech understanding.