← All models View on HuggingFace →
Qwen: Qwen3 Next 80B A3B Thinking
qwen/qwen3-next-80b-a3b-thinking
Tool useJSONReasoningStreaming
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...
Pricing
Input
$0.098 / 1M
Output
$0.78 / 1M
Specs
Context
262,144 tokens
Input
text
Output
text
Knowledge cutoff: 2025-09-30
Released: 2025-09
Supported parameters
frequency_penaltyinclude_reasoninglogit_biasmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p
Open weights · HuggingFace
34,359 downloads/mo
489 likes
apache-2.0 text-generation