Falcon-40B
Falcon-40B.
14
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Falcon-40B
Open weights
104.3
#241 of 271
Epoch Capabilities Index
- LAMBADA 77.3% #9 /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 | |
|---|---|---|---|---|
| HellaSwag | 85.3% | #10 of 74 | Epoch AI | |
| PIQA | 83% | #14 of 58 | Epoch AI | |
| LAMBADA | 77.3% | #9 of 26 | Epoch AI | |
| WinoGrande | 76.9% | #29 of 76 | Epoch AI | |
| ARC-Challenge | 61.9% | #31 of 75 | Epoch AI | |
| TriviaQA | 79.9% | #16 of 38 | Epoch AI | |
| BoolQ | 83.1% | #32 of 75 | Epoch AI | |
| OpenBookQA | 56.6% | #22 of 41 | Epoch AI | |
| MMLU | 56.9% | #90 of 127 | Epoch AI | |
| BIG-Bench Hard | 37.1% | #36 of 45 | Epoch AI | |
| AlpacaEval 2.0 | 5.61% | #37 of 40 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 33.8% | #52 of 82 | Epoch AI |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Classic | 72.9 | #6 of 17 | Official leaderboard | |
| HELM Lite | 21.7 | #67 of 71 | Official leaderboard |
Technology Innovation Institute's models over time
Technology Innovation Institute's models in color, against every other model on the Epoch Capabilities Index, by release date.