DeepSeek's flagship 1.6T MoE model with 49B active parameters, 1M-token context, MIT open weights, and world-leading coding scores at a fraction of closed-model prices.

Alibaba's flagship 2.4T MoE model with 95B active parameters, 1M-token context, native multimodal input, and top-ranked agentic performance — the largest Qwen ever.

Anthropic's frontier Opus model with 1M-token context and effort-controlled reasoning, near-Fable-5 intelligence at Opus-4.8 pricing for agents and coding.

Moonshot AI's open-weight 2.8T multimodal agentic model with 1M-token context, the world's first open 3T-class model rivaling closed frontier models.

Meta's coding-focused flagship model powering Muse Code: 1M context, higher first-attempt accuracy, reliable tool calling, and end-to-end developer workflows via the Meta Model API.