Qwen: Qwen3 Next 80B A3B Thinking coding
Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default. It’s designed for hard multi-step problems; math proofs, code synthesis/debugging, logic, and agentic...
Capabilities
Context Window 131k tokens
Max Output 32k tokens
Inputs
Outputs
Pricing (per 1M tokens)
Input $0.10
Output $0.78
Cache Read -
Cache Write -
Supported Parameters
frequency_penaltyinclude_reasoninglogit_biasmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p