Side-by-side comparison of Claude 3 Haiku (Anthropic) and Qwen 3.5 397B-A17B (Alibaba) — benchmarks, pricing, and capabilities.
| Claude 3 Haiku Anthropic | Qwen 3.5 397B-A17B Alibaba | |
|---|---|---|
| Category | LLMs | LLMs |
| Specifications | ||
| Context Window | 200K | 256K |
| Pricing (per 1M tokens) | ||
| Input Cost | $0.25 | Free |
| Output Cost | $1.25 | Free |
| Performance | ||
| Overall Score | 82.3 | 89.2 |
| ARC-Challenge | 87.1 | 91.4 |
| BigBench Hard | 72.1 | 84.0 |
| Chatbot Arena ELO | 1176.0 | 1150.0 |
| DROP | 72.5 | 84.5 |
| GSM8K | 82.5 | 91.5 |
| HumanEval | 75.0 | 87.0 |
| MATH | 58.2 | 71.0 |
| MMLU | 85.5 | 71.7 |
| TruthfulQA | 51.0 | 55.8 |
| WinoGrande | 89.4 | 78.4 |
| Community | ||
| User Rating | ★ 4.4 | ★ 4.5 |
| Reviews | 567 | 156 |