AfriqueQwen-8B: benchmark results
Provider: Other.
Aggregate Rating 1413 ± 47, rank #5507 of 10916 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SAHARA - Summarization | 17.83 | ROUGE-L | 85.1 |
| SAHARA - Title Generation | 10.13 | spBLEU-1K | 76.6 |
| SAHARA - Reading Comprehension | 36.43 | Accuracy (%) | 72.3 |
| SAHARA - Sentiment Analysis | 43.33 | Macro-F1 (%) | 63.8 |
| SAHARA - General Knowledge | 53.3 | Accuracy (%) | 55.3 |
| SAHARA - Cross-Lingual NLI | 51.9 | Accuracy (%) | 53.2 |
| SAHARA - Mathematical Word Problems | 6.8 | Exact match (%) | 53.2 |
| SAHARA - MCCR Cluster | 38.02 | Cluster score (mean of task scores) | 46.8 |
| AfriScience-MT (Zero-Shot) - English to Amharic | 33.6 | SSA-COMET (0-100; sentence-level eng->amh on the test split; | 42.9 |
| AfriScience-MT (Zero-Shot) - English to Hausa | 35.8 | SSA-COMET (0-100; sentence-level eng->hau on the test split; | 42.9 |
| SAHARA - Machine Translation (English to African) | 3.51 | spBLEU-1K | 29.8 |
| SAHARA - Overall | 19.41 | Sahara Score (mean of cluster scores) | 25.5 |
Interactive version: theaggregate.ai/model?slug=afriqueqwen-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.