Side-by-side comparison of pricing, 12 benchmarks, and generation speed.
| Metric | Hy3-preview (Reasoning) | Qwen3 1.7B (Non-reasoning) |
|---|---|---|
| Input ($/M tokens) | $0.123 | $0.11 |
| Output ($/M tokens) | $0.43 | $0.42 |
Data from Artificial Analysis API — 12 benchmarks
Qwen3 1.7B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.19/M tokens vs $0.20/M for Hy3-preview (Reasoning).
Hy3-preview (Reasoning) wins 7 out of 12 benchmarks compared to 5 for Qwen3 1.7B (Non-reasoning). See the detailed benchmark chart above for per-category results.
Qwen3 1.7B (Non-reasoning) generates tokens faster at 139 tok/s vs 117 tok/s. However, Qwen3 1.7B (Non-reasoning) has lower time-to-first-token (0.92s vs 2.40s).
Choose based on your priorities: Qwen3 1.7B (Non-reasoning) for lower cost, Hy3-preview (Reasoning) for stronger benchmark performance, and Qwen3 1.7B (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.