InternLM Intern S2

Intern S2 397B is a multimodal foundation model designed for scientific intelligence, general reasoning, and long-horizon agents. It uses a 60-layer sparse Mixture-of-Experts architecture with 512 experts and 10 activated per token, alongside 32 attention heads and 2 key-value heads. The model combines linear and full attention, with full attention every fourth layer, and supports a 262K-token context window. Its vision encoder processes image and video inputs, while large-scale reinforcement learning spans scientific domains, general reasoning, and interactive agent environments for advanced scientific problem solving and agentic workflows.