Intelligence Index
0132.8
> 53% OF MODELS
Qwen
Qwen3.5-122B-A10B
Updated: September 2026
The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of...
Specifications
- Context
- 262K
- Input Modalities
- Text, Image
- Output Modalities
- Text
- Input
- $0.29/M
- Output
- $2.4/M
Capabilities
VISIONTEXTWEBCODINGTHINKINGWRITING
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| Gemini 3.5 Flash Lite | Google | 1M | $0.3/M | $2.5/M |
| Gemini 3.5 Flash Lite | Google | 1M | $0.3/M | $2.5/M |
| Qwen3.5 Plus 2026-04-20 | Qwen | 1M | $0.3/M | $1.8/M |
| Minimax M2.7 | Minimax | 205K | $0.3/M | $1.2/M |
| Minimax M2.7 | Minimax | 205K | $0.3/M | $1.2/M |
Performance Metrics
Coding Index
0245.7
> 47% OF MODELS
Agentic Index
0321.3
> 42% OF MODELS
Average Response Performance
- Output Speed
- 131 tok/s
- Time To First Token
- 2.3s
- Time To First Answer Token
- 17.58s
- End To End Response Time
- 21.39s
DATA SOURCE: Artificial Analysis