DeepSeek R1 Distill Llama 70B vs DeepSeek V3.2 Speciale
DeepSeek vs DeepSeek — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 70.1)
- •DeepSeek R1 Distill Llama 70B is cheaper ($0.10 vs $0.29 per 1M input)
- •DeepSeek V3.2 Speciale has a larger context window (164K)
| DeepSeek R1 Distill Llama 70B | DeepSeek V3.2 Speciale | |
|---|---|---|
| Intelligence index | 70.1 | 89.9 |
| Developer | DeepSeek | DeepSeek |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 128,000 tokens | 163,840 tokens |
| Input price | $0.10 / 1M | $0.29 / 1M |
| Output price | $0.40 / 1M | $0.43 / 1M |
| Speed | 37 tok/s | — |
| Released | January 20, 2025 | December 1, 2025 |
| Parameters | 70600000000 | — |
| Input modalities | — | Text |
| Output modalities | — | Text |
Shared benchmarks
DeepSeek R1 Distill Llama 70B
DeepSeek V3.2 Speciale
AIME 2025
53.7
96.7
GPQA Diamond
65.2
87.1
Humanity’s Last Exam
6.1
26.1
LiveCodeBench
57.5
89.6
MMLU-Pro
79.5
86.3