Side-by-side comparison of Claude 3.5 Sonnet (Anthropic) and Imagen 3 (Google) — benchmarks, pricing, and capabilities.
| Claude 3.5 Sonnet Anthropic | Imagen 3 Google | |
|---|---|---|
| Category | LLMs | Image Generation |
| Specifications | ||
| Context Window | 200K | — |
| Pricing (per 1M tokens) | ||
| Input Cost | $3.00 | — |
| Output Cost | $15.00 | — |
| Performance | ||
| Overall Score | 91.2 | 87.0 |
| ARC-Challenge | 96.2 | — |
| BigBench Hard | 86.0 | — |
| Chatbot Arena ELO | 1104.0 | — |
| DPG-Bench | — | 80.0 |
| DROP | 86.5 | — |
| GSM8K | 78.3 | — |
| GenEval | — | 82.5 |
| HumanEval | 90.2 | — |
| MATH | 75.2 | — |
| MMLU | 88.1 | — |
| TruthfulQA | 71.0 | — |
| WinoGrande | 82.0 | — |
| Community | ||
| User Rating | ★ 4.6 | ★ 4.4 |
| Reviews | 890 | 560 |