Side-by-side comparison of pricing, 12 benchmarks, and generation speed.
| Metric | Granite 3.3 8B (Non-reasoning) | Mercury 2 |
|---|---|---|
| Input ($/M tokens) | $0.03 | $0.25 |
| Output ($/M tokens) | $0.25 | $0.75 |
Data from Artificial Analysis API — 12 benchmarks
Granite 3.3 8B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.09/M tokens vs $0.38/M for Mercury 2.
Mercury 2 wins 7 out of 12 benchmarks compared to 5 for Granite 3.3 8B (Non-reasoning). See the detailed benchmark chart above for per-category results.
Mercury 2 generates tokens faster at 753 tok/s vs 702 tok/s. However, Mercury 2 has lower time-to-first-token (3.49s vs 18.45s).
Choose based on your priorities: Granite 3.3 8B (Non-reasoning) for lower cost, Mercury 2 for stronger benchmark performance, and Mercury 2 for faster generation. For latency-sensitive apps, check the TTFT comparison above.