INTELLECT-2 model card (Hugging Face) 11 May 2025
- AIME24 78.8
- AIME25 64.9
- LiveCodeBench (v5) 67.8
- GPQA-Diamond 66.8
- IFEval 81.5
INTELLECT-2
INTELLECT-2.
5
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Prime Intellect May 2025
INTELLECT-2
Open weights Flagship
No Epoch Capabilities Index score yet
- LiveCodeBench v5 67.8% #14 /44
- AIME 2025 64.9% #120 /143
- IFEval 81.5% #93 /126
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 | |
|---|---|---|---|---|
| IFEval | 81.5% | #93 of 126 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2024 | 78.8% | #45 of 87 | Model card | |
| GPQA Diamond | 66.8% | #183 of 284 | Model card | |
| AIME 2025 | 64.9% | #120 of 143 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v5 | 67.8% | #14 of 44 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it.
INTELLECT-2: A Reasoning Model Trained Through Globally Decentralized Reinforcement Learning 12 May 2025
Capabilities
- AIME 2024 78.8
- AIME 2025 64.9
- LiveCodeBench v5 67.8
- GPQA Diamond 66.8
- IFEval 81.5