Side-by-side comparison of pricing, 12 benchmarks, and generation speed.
| Metric | Qwen3 4B 2507 (Reasoning) | gpt-oss-20b (low) |
|---|---|---|
| Input ($/M tokens) | $0 | $0.06 |
| Output ($/M tokens) | $0 | $0.2 |
Data from Artificial Analysis API — 12 benchmarks
Qwen3 4B 2507 (Reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $0.10/M for gpt-oss-20b (low).
It's a tie — both models win 5 benchmarks each across 12 evaluated categories. See the detailed benchmark chart above for per-category results.
gpt-oss-20b (low) generates tokens faster at 226 tok/s vs 0 tok/s. Qwen3 4B 2507 (Reasoning) also has lower time-to-first-token (0.00s vs 0.50s).
Choose based on your priorities: Qwen3 4B 2507 (Reasoning) for lower cost, both perform similarly on benchmarks, and gpt-oss-20b (low) for faster generation. For latency-sensitive apps, check the TTFT comparison above.