Side-by-side comparison of Qwen 3.5 397B-A17B (Alibaba) and Claude Agent SDK (Anthropic) — benchmarks, pricing, and capabilities.
| Qwen 3.5 397B-A17B Alibaba | Claude Agent SDK Anthropic | |
|---|---|---|
| Category | LLMs | Skills & Tools |
| Specifications | ||
| Context Window | 256K | — |
| Pricing (per 1M tokens) | ||
| Input Cost | Free | — |
| Output Cost | Free | — |
| Performance | ||
| Overall Score | 89.2 | 86.5 |
| ARC-Challenge | 89.8 | — |
| BigBench Hard | 84.0 | — |
| Chatbot Arena ELO | 1197.0 | — |
| DROP | 84.5 | — |
| GSM8K | 91.5 | — |
| HumanEval | 87.0 | — |
| MATH | 71.0 | — |
| MMLU | 74.5 | — |
| TruthfulQA | 63.0 | — |
| WinoGrande | 81.5 | — |
| Community | ||
| User Rating | ★ 4.5 | ★ 4.4 |
| Reviews | 156 | 185 |