Devstral 2 vs Qwen3.6 Max
Mistral AI vs Alibaba — benchmarks, pricing, and capabilities side by side.
- •Qwen3.6 Max has the higher intelligence index (88.8 vs 54.3)
- •Devstral 2 is cheaper ($0.40 vs $1.04 per 1M input)
- •Devstral 2 is faster
| Devstral 2 | Qwen3.6 Max | |
|---|---|---|
| Intelligence index | 54.3 | 88.8 |
| Developer | Mistral AI | Alibaba |
| Type | LLM | LLM |
| Access | Open weights | API only |
| Context window | 262,144 tokens | 262,144 tokens |
| Input price | $0.40 / 1M | $1.04 / 1M |
| Output price | $2.00 / 1M | $6.24 / 1M |
| Speed | 51 tok/s | 36 tok/s |
| Released | December 9, 2025 | April 27, 2026 |
| Parameters | — | — |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
Shared benchmarks
Devstral 2
Qwen3.6 Max
GPQA Diamond
59.4
88.8
Humanity’s Last Exam
3.6
28.9