Qwen3.5-0.8B-Base: benchmark results
Qwen3.5 0.8B base checkpoint. Provider: Alibaba. Released 2026-03-02. Access: Open.
Aggregate Rating 1322 ± 26, rank #8299 of 10916 rated models, from 279 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Finnish NLU - Tydiqa FI | 66.76 | Character F1 (%) | 75.6 |
| EuroEval German NLU - Germanquad | 56.78 | Character F1 (%) | 73.1 |
| EuroEval Polish NLU - PoQuAD | 49.42 | Character F1 (%) | 69.5 |
| EuroEval Bulgarian NLU - Cinexio | 46.31 | Matthews correlation coefficient (%) | 68.7 |
| EuroEval Spanish NLU - MLQA ES | 61.66 | Character F1 (%) | 61.4 |
| EuroEval Bulgarian NLU - MultiWikiQA BG | 66.55 | Character F1 (%) | 58.8 |
| EuroEval English NLU - SQuAD | 81.16 | Character F1 (%) | 58.7 |
| EuroEval Slovak NLU - Multi Wiki QA SK | 60.1 | Character F1 (%) | 57.7 |
| EuroEval Greek NLU - MultiWikiQA EL | 58.74 | Character F1 (%) | 56.7 |
| EuroEval Albanian NLU - MultiWikiQA SQ | 51.86 | Character F1 (%) | 55.3 |
| EuroEval Italian NLU - SQuAD IT | 67.52 | Character F1 (%) | 55.3 |
| EuroEval Bosnian NLU - Multi Wiki QA BS | 62.51 | Character F1 (%) | 55.2 |
Interactive version: theaggregate.ai/model?slug=qwen3-5-0-8b-base · How It Works · Data refreshed daily, snapshot 2026-11-08.