Llama 3.1 70B Instruct vs Sonar Reasoning Pro
Meta vs Perplexity — benchmarks, pricing, and capabilities side by side.
- •Sonar Reasoning Pro has the higher intelligence index (95.7 vs 56)
- •Llama 3.1 70B Instruct is cheaper ($0.40 vs $2.00 per 1M input)
- •Llama 3.1 70B Instruct has a larger context window (131K)
| Llama 3.1 70B Instruct | Sonar Reasoning Pro | |
|---|---|---|
| Intelligence index | 56 | 95.7 |
| Developer | Meta | Perplexity |
| Type | LLM | Multimodal |
| Access | Open weights | API only |
| Context window | 131,072 tokens | 128,000 tokens |
| Input price | $0.40 / 1M | $2.00 / 1M |
| Output price | $0.40 / 1M | $8.00 / 1M |
| Speed | 1204 tok/s | — |
| Released | July 23, 2024 | March 7, 2025 |
| Parameters | — | — |
| Input modalities | Text | Text, Image |
| Output modalities | Text | Text |
Shared benchmarks
Llama 3.1 70B Instruct
Sonar Reasoning Pro
MATH-500
64.9
95.7