Cerebras GPT 13B — Benchmarks
Benchmark scores for Cerebras GPT 13B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Knowledge
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| HellaSwag | 59.4% | #36 / 42 | #63 / 76 |
| ARC Challenge | 32.4% | #48 / 51 | #73 / 77 |
| BoolQ | 61.1% | #32 / 33 | #70 / 77 |
| OpenBookQA | 35.8% | #19 / 21 | #37 / 42 |
| MMLU | 26.2% | #75 / 77 | #132 / 136 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| WinoGrande | 60.8% | #36 / 46 | #70 / 80 |
| PIQA | 73.5% | #31 / 35 | #54 / 60 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.