Efficient Qwen thinking model for local reasoning, math, and coding agents
Details
| Field | Value |
|---|---|
| id | qwen3-next-80b-a3b-thinking |
| family | qwen |
| modality | text->text |
| context_length | 131072 |
| max_output_tokens | 32768 |
| knowledge_cutoff | 2025-04 |
| release_date | 2025-09 |
| is_open_weights | yes |
| supports_tool_call | yes |
| supports_reasoning | yes |
| supports_structured_output | no |
| supports_attachment | no |
| input_modalities | text |
| output_modalities | text |