DeepSeek V3.2 Speciale vs Mixtral 8x7B Instruct
DeepSeek vs Mistral AI — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 26.1)
- •DeepSeek V3.2 Speciale is cheaper ($0.29 vs $0.50 per 1M input)
| DeepSeek V3.2 Speciale | Mixtral 8x7B Instruct | |
|---|---|---|
| Intelligence index | 89.9 | 26.1 |
| Developer | DeepSeek | Mistral AI |
| Type | LLM | LLM |
| Access | Open weights | — |
| Context window | 163,840 tokens | — |
| Input price | $0.29 / 1M | $0.50 / 1M |
| Output price | $0.43 / 1M | $0.70 / 1M |
| Speed | — | — |
| Released | December 1, 2025 | December 11, 2023 |
| Parameters | — | — |
| Input modalities | Text | — |
| Output modalities | Text | — |
Shared benchmarks
DeepSeek V3.2 Speciale
Mixtral 8x7B Instruct
GPQA Diamond
87.1
29.2
Humanity’s Last Exam
26.1
4.5
LiveCodeBench
89.6
6.6
MMLU-Pro
86.3
38.7