Side-by-side comparison of pricing, 12 benchmarks, and generation speed.
| Metric | GPT-5.4 mini (Non-Reasoning) | Claude 4.5 Haiku (Reasoning) |
|---|---|---|
| Input ($/M tokens) | $0.75 | $1 |
| Output ($/M tokens) | $4.5 | $5 |
Data from Artificial Analysis API — 12 benchmarks
GPT-5.4 mini (Non-Reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $1.69/M tokens vs $2.00/M for Claude 4.5 Haiku (Reasoning).
Claude 4.5 Haiku (Reasoning) wins 11 out of 12 benchmarks compared to 0 for GPT-5.4 mini (Non-Reasoning). See the detailed benchmark chart above for per-category results.
GPT-5.4 mini (Non-Reasoning) generates tokens faster at 202 tok/s vs 139 tok/s. GPT-5.4 mini (Non-Reasoning) also has lower time-to-first-token (0.45s vs 9.86s).
Choose based on your priorities: GPT-5.4 mini (Non-Reasoning) for lower cost, Claude 4.5 Haiku (Reasoning) for stronger benchmark performance, and GPT-5.4 mini (Non-Reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.