DeepSeek V3.2 Speciale vs Llama 3.1 Tulu3 405B
DeepSeek vs Allen Institute for AI — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 57.5)
- •Llama 3.1 Tulu3 405B is cheaper ($0.00 vs $0.29 per 1M input)
| DeepSeek V3.2 Speciale | Llama 3.1 Tulu3 405B | |
|---|---|---|
| Intelligence index | 89.9 | 57.5 |
| Developer | DeepSeek | Allen Institute for AI |
| Type | LLM | LLM |
| Access | Open weights | — |
| Context window | 163,840 tokens | — |
| Input price | $0.29 / 1M | $0.00 / 1M |
| Output price | $0.43 / 1M | $0.00 / 1M |
| Speed | — | — |
| Released | December 1, 2025 | January 30, 2025 |
| Parameters | — | — |
| Input modalities | Text | — |
| Output modalities | Text | — |
Shared benchmarks
DeepSeek V3.2 Speciale
Llama 3.1 Tulu3 405B
GPQA Diamond
87.1
51.6
Humanity’s Last Exam
26.1
3.5
LiveCodeBench
89.6
29.1
MMLU-Pro
86.3
71.6