Mercury
Mercury.
9
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Inception Labs Jun 2025
Mercury
Closed weights Flagship
121.2
#337 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
- SciCode 18% #48 /49
- MMLU-Pro 69% #126 /167
- MATH-500 83% #59 /70
Where it ranks
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.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMArena (Elo) | 1,305 | #226 of 335 | Official leaderboard | |
| MMLU-Pro | 69% | #126 of 167 | Lab model card | |
| Humanity's Last Exam | 3.40% | #154 of 154 | Lab model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 51% | #242 of 302 | Lab model card | |
| MATH-500 | 83% | #59 of 70 | Lab model card | |
| AIME 2024 | 30% | #79 of 88 | Lab model card |
Coding
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 32 | #29 of 188 | Official leaderboard |
Its lab's numbers put it at 121.2, number 337 of 404
The Self-Reported Index fits one score to the 7 capability results Inception Labs reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 118.9 to 123.6. How we compute it.
- HumanEval 85
- MATH-500 83
- MMLU-Pro 69
- GPQA Diamond 51
- AIME 2024 30
- SciCode 18
- Humanity's Last Exam 3.4
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.