1T / 32B (MoE)
Modified MIT
Thinking agent with step-by-step reasoning and dynamic tool use; 256K context; SOTA on HLE (44.9% w/ tools) and BrowseComp (60.2%); 200-300 sequential tool calls; native INT4 quantization for 2x speed; excels at agentic coding/workflows; tops SWE-Bench Verified (71.3%).