| Note | 14B-parameter open-weight (MIT) flagship of Microsoft's Phi small-model family; top-of-class STEM reasoning for its size but limited to a 16K context window. Available on Azure AI Foundry, HuggingFace, and Ollama. | Current flagship general-purpose model; strong coding, reasoning, writing, and agentic tasks. Tiered pricing above a context threshold. | First publicly available Mythos-class model (above Opus); Mythos 5 weights with safety classifiers active. 1M context, $10/$50, highest benchmarks of any GA model. Refusals return stop_reason refusal; mandatory 30-day retention. |
|---|
