Introducing Mercury 2 (Inception blog) 24 Feb 2026
- GPQA Diamond 74
- SciCode 38
- IFBench 71
Mercury 2.
Closed weights
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMArena (Elo) | 1,343 | #185 of 333 | Official leaderboard | |
| IFBench | 71% | #39 of 70 | Model card | |
| Vectara Hallucination Leaderboard (2025 dataset) | 12.3% | #84 of 104 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 74% | #147 of 284 | Model card | |
| CritPt | 0.85% | #100 of 160 | Epoch AI |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WeirdML | 43.2% | #66 of 112 | Official leaderboard | |
| ALE-Bench | 786 | #72 of 121 | Epoch AI | |
| SciCode | 38% | #33 of 47 | Model card | |
| Code Arena | 1,170 | #121 of 124 | Official leaderboard | |
| WebDev Arena | 1,167 | #103 of 105 | Epoch AI |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 28 | #44 of 183 | Official leaderboard | |
| PinchBench 2.0 | 39.5% | #52 of 57 | Official leaderboard |
The benchmarks the lab chose to headline, with the scores it gave.