Babel-9B-Chat: benchmark results
Provider: Other.
Aggregate Rating 1491 ± 34, rank #2499 of 10916 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SAHARA - Paraphrase | 31.33 | spBLEU-1K | 83 |
| FilBench - Classical NLP | 87.67 | Category Score (%) | 61.1 |
| FilBench - Cultural Knowledge | 60.06 | Category Score (%) | 51.4 |
| FilBench - Reading Comprehension | 56.49 | Category Score (%) | 51.4 |
| SAHARA - Machine Translation (French to African) | 1.19 | spBLEU-1K | 51.1 |
| SAHARA - News Classification | 24.06 | Macro-F1 (%) | 46.8 |
| FilBench - Generation | 6.79 | Category Score (%) | 43.6 |
| SAHARA - General Knowledge | 44.51 | Accuracy (%) | 42.6 |
| FilBench | 52.75 | FilBench Score (%) | 41.7 |
| SAHARA - Machine Translation (English to African) | 4.37 | spBLEU-1K | 40.4 |
| SAHARA - Text Generation Cluster | 9.35 | Cluster score (mean of task scores) | 38.3 |
| SAHARA - Machine Translation (African to African) | 0.55 | spBLEU-1K | 37.2 |
Interactive version: theaggregate.ai/model?slug=babel-9b-chat · How It Works · Data refreshed daily, snapshot 2026-11-08.