Llama 3.1 Tulu3 405B vs Sonar Reasoning Pro
Allen Institute for AI vs Perplexity — benchmarks, pricing, and capabilities side by side.
- •Sonar Reasoning Pro has the higher intelligence index (95.7 vs 57.5)
- •Llama 3.1 Tulu3 405B is cheaper ($0.00 vs $2.00 per 1M input)
| Llama 3.1 Tulu3 405B | Sonar Reasoning Pro | |
|---|---|---|
| Intelligence index | 57.5 | 95.7 |
| Developer | Allen Institute for AI | Perplexity |
| Type | LLM | Multimodal |
| Access | — | API only |
| Context window | — | 128,000 tokens |
| Input price | $0.00 / 1M | $2.00 / 1M |
| Output price | $0.00 / 1M | $8.00 / 1M |
| Speed | — | — |
| Released | January 30, 2025 | March 7, 2025 |
| Parameters | — | — |
| Input modalities | — | Text, Image |
| Output modalities | — | Text |
Shared benchmarks
Llama 3.1 Tulu3 405B
Sonar Reasoning Pro
MATH-500
77.8
95.7