DeepSeek R1 is here: Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass.

An open‑source Mixture‑of‑Experts chat model with a massive 128 K token context window, optimized for coding, math, reasoning, and long‑form workflows at low cost.

Streamlined for quick, casual chat and basic reasoning with the signature X-style tone.

A general-purpose chat assistant optimized for informal dialogue and developer-friendly interactions.

A high-throughput, multimodal model with a massive 1 M‑token context window, optimized for real-time, large‑scale applications.

Expert-level multimodal LLM with advanced “Deep Think” reasoning—designed for complex coding, math, and scientific workflows.

Cost-effective multimodal workhorse with adaptive reasoning, high throughput, and built-in chain-of-thought.

Anthropic’s flagship—the most powerful model yet for sustained coding, deep reasoning, and agentic workflows.

Earlier hybrid model with instant or extended reasoning, suitable for general-purpose tasks and coding.

Earlier hybrid model with instant or extended reasoning, suitable for general-purpose tasks and coding.