Side-by-side comparison of Llama 3.1 405B (Meta) and Claude 3.5 Haiku (Anthropic) — benchmarks, pricing, and capabilities.
| Llama 3.1 405B Meta | Claude 3.5 Haiku Anthropic | |
|---|---|---|
| Category | LLMs | LLMs |
| Specifications | ||
| Context Window | 128K | 200K |
| Pricing (per 1M tokens) | ||
| Input Cost | Free | $0.80 |
| Output Cost | Free | $4.00 |
| Performance | ||
| Overall Score | 88.4 | 84.0 |
| ARC-Challenge | 87.6 | — |
| BigBench Hard | 82.3 | 78.0 |
| Chatbot Arena ELO | 1126.0 | — |
| DROP | 82.0 | — |
| GSM8K | 91.0 | 86.5 |
| HumanEval | 85.0 | 80.5 |
| MATH | 70.1 | 65.0 |
| MMLU | 84.2 | 81.5 |
| TruthfulQA | 65.2 | — |
| WinoGrande | 79.4 | — |
| Community | ||
| User Rating | ★ 4.4 | ★ 4.4 |
| Reviews | 420 | 290 |