Call any LLM using the OpenAI format. Supports 100+ providers with load balancing, fallbacks, and spend tracking built in.
TypeScript toolkit for building AI-powered apps. Unified interface for text generation, streaming, and structured outputs across any provider.
Unified API for 200+ LLMs. Route requests to the best model for cost, speed, or capability with an OpenAI-compatible interface.
AI gateway with observability, guardrails, and fallbacks. Routes traffic across providers with cost and latency tracking.
Google DeepMind flagship model with 1M token context, native multimodality, and deep reasoning capabilities.
Meta latest open-weights model. State-of-the-art among open models, available in 8B and 70B parameter variants.
Alibaba open-source model family with strong multilingual and coding performance. Ranges from 0.5B to 72B parameters.
The original Mistral open model. Outperforms Llama 2 13B on most benchmarks while being faster and more efficient.
xAI most capable model with real-time X/Twitter data access, strong reasoning, and a massive context window.
High-throughput and memory-efficient inference and serving engine for LLMs. PagedAttention + continuous batching.
Run large language models locally. Supports Llama, Mistral, Gemma and dozens of others with a simple CLI and API.
Mistral AI's top-tier model with strong reasoning, multilingual support, and function calling.
Anthropic's most intelligent model, excelling at complex reasoning, coding, and agentic tasks.
OpenAI's flagship multimodal model. Supports text, image, and audio natively with improved speed and cost.