Mercury 3 vs SWE-2: Specs & Benchmark Comparison
Mercury 3 is developed by Inception, while SWE-2 comes from Cognition. SWE-2 was released in September 2026, and Mercury 3 followed a month later in October 2026.
We have 12 benchmark results for Mercury 3 and 4 benchmark results for SWE-2, but they were measured on different benchmarks, so there is no like-for-like scoreboard.
Mercury 3 is available through an API at $0.10 per million input tokens with a 1M-token context window. We do not track a hosted API for SWE-2, so pricing and context cannot be compared directly.
| Characteristic | Mercury 3 | SWE-2 |
|---|---|---|
| Company | Inception | Cognition |
| Release Date | October 19, 2026 | September 10, 2026 |
| Parameters | — | — |
| Multimodal | No | No |
| Context (input) | 1.0M | — |
| Context (output) | 131K | — |
| Input Price / 1M | $0.10 | — |
| Output Price / 1M | $0.35 | — |
| Average Score | 60.1% | 60.8% |
Verdict
Mercury 3 leads in 1 out of 2 comparison categories.
Both models show comparable average scores: Mercury 3 — 0.6, SWE-2 — 0.6.
Mercury 3 is newer: released 10/19/2026 vs 9/10/2026.
More About These Models
Related Comparisons
Frequently Asked Questions
Which is better for coding — Mercury 3 or SWE-2?
Which model is cheaper — Mercury 3 or SWE-2?
Which has a larger context window — Mercury 3 or SWE-2?
The Mercury 3 and SWE-2 comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the Mercury 3 or SWE-2 page. See also the complete list of AI model comparisons.