Open Mixture of Experts LLM (230B, 10B active) for reasoning, coding, and tool-use/agent workflows
DeepSeek-V3.1: hybrid inference LLM with Think/Non-Think modes, stronger agents, 128K context, strict function calling.
State-of-the-art open model for reasoning, code, math, and tool calling - suitable for edge agents
Natural and expressive voices in multiple languages. For voice agents and brand ambassadors.
Latency-optimized language model excelling in code, math, general knowledge, and instruction-following.