StableBeluga2 — Benchmarks
Benchmark scores for StableBeluga2 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
No overall rank for StableBeluga2: an overall score is only published when a model has been measured on enough of the benchmarks current models are still submitted to. Its individual scores below stand on their own — see the methodology for how the overall score is built.
Knowledge
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| ARC Challenge | 86.1% | #8 / 56 | #10 / 76 |
| BoolQ | 89.4% | #1 / 38 | #5 / 76 |
| MMLU | 68.6% | #33 / 85 | #71 / 136 |
| HellaSwag | 84.1% | #9 / 47 | #14 / 75 |
Math
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| GSM8K | 69.6% | #19 / 62 | #27 / 93 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| BIG-Bench Hard | 69.3% | #9 / 38 | #12 / 50 |
| PIQA | 83.3% | #8 / 40 | #12 / 59 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.