Frontier AI models: Claude 3.5 Sonnet and Haiku with superior reasoning and coding.

Multimodal frontier models: Gemini 1.5 Pro & Flash with 2M token context windows.

LPU Inference Engine delivering 500+ tokens/second for open-weight LLMs.

Industry-leading frontier models: GPT-4o, o1, embeddings, and Whisper speech.

A unified API interface to all major AI models with automatic fallbacks and lowest pricing.