text-ada-001: benchmark results
OpenAI's text-ada-001, the smallest and fastest model in the Ada/Babbage/Curie/Davinci GPT-3 lineup, fine-tuned with human feedback for instruction-following text completion. Provider: OpenAI. Released 2022-01-27. Access: API.
Aggregate Rating 1005 ± 30, rank #1655 of 1656 rated models, from 35 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RealToxicityPrompts | 0.04 | Toxic fraction | 68.3 |
| HELM Classic - LSAT | 21.3 | Exact Match (%) | 66.9 |
| HELM Classic - BBQ | 35.9 | Exact Match (%) | 52.4 |
| HELM Classic - TruthfulQA | 23.19 | Exact Match (%) | 49.2 |
| HELM Classic - Entity Matching | 80.85 | Exact Match (%) | 48.5 |
| HELM Classic - CNN/DailyMail | 13.56 | ROUGE-2 (%) | 45.1 |
| HELM Classic - LegalSupport | 51.47 | Exact Match (%) | 41.9 |
| HELM Classic - Synthetic Reasoning Natural | 14.53 | F1 (%) | 41.2 |
| HELM Classic - XSUM | 3.44 | ROUGE-2 (%) | 12.2 |
| HELM Classic - IMDB | 82.2 | Exact Match (%) | 12.1 |
| HELM Classic - CivilComments | 50.33 | Exact Match (%) | 10.6 |
| HELM Classic - MATH | 2.04 | Equivalent (%) | 10.3 |
Interactive version: theaggregate.ai/model?slug=text-ada-001 · How It Works · Data refreshed daily, snapshot 2026-10-20.