Intelligence Index
0114.4
> 27% OF MODELS
Qwen
Qwen3 30B A3B Thinking 2507
Updated: July 2026
Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for “thinking mode,” where internal reasoning traces are separated...
Specifications
- Context
- 82K
- Input
- $0.2/M
- Output
- $2.4/M
Capabilities
TEXTWEBCODINGTHINKINGWRITING
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| Chatgpt 5.4 Nano | OpenAI | 400K | $0.2/M | $1.25/M |
| Chatgpt 5.4 Nano | OpenAI | 400K | $0.2/M | $1.25/M |
| Ministral 3 14B 2512 | MistralAI | 262K | $0.2/M | $0.2/M |
| Qwen3 VL 30B A3B Thinking | Qwen | 262K | $0.2/M | $2.4/M |
| Qwen3 VL | Qwen | 262K | $0.2/M | $0.88/M |
Performance Metrics
Coding Index
0212.1
> 10% OF MODELS
Agentic Index
031.8
> 13% OF MODELS
DATA SOURCE: Artificial Analysis