GPT-3 175B (davinci)
GPT-3 175B (davinci).
9
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
OpenAI May 2020
GPT-3 175B (davinci)
Closed weights
No Epoch Capabilities Index score yet
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 | 77.5% | #4 of 12 | Stanford HELM, via Epoch AI | |
| BoolQ | 76.4% | #3 of 3 | Official leaderboard | |
| CB | 75.6% | #3 of 3 | Official leaderboard | |
| MultiRC | 75.4% | #3 of 3 | Official leaderboard | |
| SuperGLUE | 71.8% | #3 of 3 | Official leaderboard | |
| WSC | 80.1% | #3 of 3 | Official leaderboard | |
| PIQA | 82.8% | #1 of 1 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 9.00% | #14 of 24 | Stanford HELM, via Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| METR Time Horizon | 0.1 min | #25 of 26 | Claim in metr.org, via Epoch AI |