LLaMA-33B
LLaMA-33B.
Meta Released 24 Feb 2023 11
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Meta Feb 2023 LLaMA-33B
Open weights
107.3
#234 of 271
Epoch Capabilities Index
- LAMBADA 77.2% #10 /26
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 | |
|---|---|---|---|---|
| TriviaQA | 83.8% | #9 of 38 | Epoch AI | |
| HellaSwag | 82.8% | #18 of 74 | Epoch AI | |
| BoolQ | 86.1% | #19 of 75 | Epoch AI | |
| ARC-Challenge | 67.5% | #27 of 75 | Epoch AI | |
| PIQA | 82.3% | #22 of 58 | Epoch AI | |
| LAMBADA | 77.2% | #10 of 26 | Epoch AI | |
| OpenBookQA | 58.6% | #16 of 41 | Epoch AI | |
| WinoGrande | 76% | #32 of 76 | Epoch AI | |
| BIG-Bench Hard | 50% | #23 of 45 | Epoch AI | |
| MMLU | 58.7% | #85 of 127 | Epoch AI |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 44.1% | #43 of 82 | Epoch AI |
Meta's models over time
Meta's models in color, against every other model on the Epoch Capabilities Index, by release date.