Introducing Mercury 2 (Inception blog) Feb 24, 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 | |
|---|---|---|---|---|
| IFBench | 71% | #40 of 80 | Lab model card | |
| LMArena (Elo) | 1,343 | #187 of 335 | Official leaderboard | |
| StudyArena Text | 1,222 | #52 of 85 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 12.3% | #84 of 104 | Official leaderboard | |
| DesignArena ASCII Art | 981 | #77 of 77 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 74% | #156 of 302 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WeirdML | 43.2% | #66 of 112 | Official leaderboard | |
| SciCode | 38% | #33 of 49 | Lab model card | |
| ALE-Bench | 1,190 | #86 of 126 | Official leaderboard | |
| DesignArena Website | 1,011 | #131 of 151 | Official leaderboard | |
| DesignArena Code | 997 | #129 of 143 | Official leaderboard | |
| DesignArena UI Component | 970 | #126 of 139 | Official leaderboard | |
| DesignArena Game Dev | 963 | #131 of 143 | Official leaderboard | |
| DesignArena 3D Design | 956 | #125 of 136 | Official leaderboard | |
| DesignArena SVG | 991 | #95 of 103 | Official leaderboard | |
| DesignArena Data Visualization | 1,012 | #112 of 118 | Official leaderboard | |
| Code Arena | 1,171 | #125 of 128 | Official leaderboard | |
| WebDev Arena | 1,167 | #103 of 105 | arena.ai, via Epoch AI |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 28 | #49 of 188 | Official leaderboard | |
| PinchBench 2.0 | 39.5% | #52 of 57 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| nocot-bench (NCRI) | 55.1 | #155 of 185 | Official leaderboard |
The benchmarks the lab chose to headline, with the scores it gave.