Intelligence Index
0116.9
> 27% OF MODELS
Qwen
Qwen3 Next 80B A3B Thinking
Updated: September 2026
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...
Specifications
- Context
- 262K
- Input Modalities
- Text
- Output Modalities
- Text
- Input
- $0.15/M
- Output
- $1.2/M
Capabilities
TEXTWEBCODINGTHINKINGWRITING
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| Qwen3.8 Flash | Qwen | 1M | $0.15/M | $0.47/M |
| Z.ai: GLM 5.3 Flash (batch) | Z-AI | 1M | $0.15/M | $0.5/M |
| Qwen3.6 35B A3B | Qwen | 262K | $0.15/M | $1/M |
| Mistral Small 4 | MistralAI | 262K | $0.15/M | $0.6/M |
| Ministral 3 8B 2512 | MistralAI | 262K | $0.15/M | $0.15/M |
Performance Metrics
Coding Index
0217.4
> 16% OF MODELS
Agentic Index
032.1
> 13% OF MODELS
DATA SOURCE: Artificial Analysis