Side-by-side comparison of Claude 3.5 Haiku (Anthropic) and GPT-4o (OpenAI) — benchmarks, pricing, and capabilities.
| Claude 3.5 Haiku Anthropic | GPT-4o OpenAI | |
|---|---|---|
| Category | LLMs | LLMs |
| Specifications | ||
| Context Window | 200K | 128K |
| Pricing (per 1M tokens) | ||
| Input Cost | $0.80 | $2.50 |
| Output Cost | $4.00 | $10.00 |
| Performance | ||
| Overall Score | 84.0 | 92.5 |
| ARC-Challenge | — | 96.9 |
| BigBench Hard | 78.0 | 87.2 |
| Chatbot Arena ELO | — | 1270.0 |
| DROP | — | 88.1 |
| GSM8K | 86.5 | 95.3 |
| HumanEval | 80.5 | 92.0 |
| MATH | 65.0 | 76.6 |
| MMLU | 81.5 | 88.7 |
| TruthfulQA | — | 72.2 |
| WinoGrande | — | 89.6 |
| Community | ||
| User Rating | ★ 4.4 | ★ 4.7 |
| Reviews | 290 | 1240 |