Intelligence Index
017.2
> 11% OF MODELS
Qwen
Qwen3 32B
Updated: September 2026
Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for...
Specifications
- Context
- 131K
- Input Modalities
- Text
- Output Modalities
- Text
- Input
- $0.08/M
- Output
- $0.28/M
Capabilities
TEXTWEBCODINGTHINKINGWRITING
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| Gemma 3 27B | Google | 131K | $0.08/M | $0.45/M |
| Z.ai: GLM 5.3 Flash (batch) | Z-AI | 1M | $0.075/M | $0.25/M |
| Mistral Small 4 (batch) | MistralAI | 262K | $0.075/M | $0.3/M |
| ByteDance Seed: Seed 1.6 Flash | ByteDance | 262K | $0.075/M | $0.3/M |
| Ministral 3 8B 2512 (batch) | MistralAI | 262K | $0.075/M | $0.075/M |
Performance Metrics
Coding Index
0215.3
> 12% OF MODELS
Agentic Index
030.9
> 8% OF MODELS
Average Response Performance
- Output Speed
- 105.3 tok/s
- Time To First Token
- 2.45s
- Time To First Answer Token
- 21.44s
- End To End Response Time
- 26.19s
DATA SOURCE: Artificial Analysis