ada (350M): benchmark results
Provider: Other.
Aggregate Rating 1032 ± 37, rank #10911 of 10916 rated models, from 32 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RealToxicityPrompts | 0.05 | Toxic fraction | 48.8 |
| HELM Classic - BLiMP | 81.77 | Exact Match (%) | 45.2 |
| HELM Classic - TruthfulQA | 21.51 | Exact Match (%) | 37.1 |
| HELM Classic - LSAT | 18.84 | Exact Match (%) | 29.4 |
| HELM Classic - Synthetic Reasoning Natural | 8.83 | F1 (%) | 29.4 |
| HELM Classic - CivilComments | 51.67 | Exact Match (%) | 25.8 |
| HELM Classic - CNN/DailyMail | 9.02 | ROUGE-2 (%) | 19.5 |
| HELM Classic - IMDB | 84.9 | Exact Match (%) | 18.2 |
| HELM Classic - Dyck | 40.6 | Exact Match (%) | 17.6 |
| HELM Classic - BBQ | 32.17 | Exact Match (%) | 17.1 |
| HELM Classic - MATH | 4.64 | Equivalent (%) | 14.7 |
| HELM Classic - MMLU | 24.28 | Exact Match (%) | 11.4 |
Interactive version: theaggregate.ai/model?slug=ada-350m · How It Works · Data refreshed daily, snapshot 2026-11-08.