Qwen3.8 2.4T A95B (xHigh): benchmark results
Provider: Alibaba. Released 2026-08-12. Access: Open.
Aggregate Rating 1807 ± 22, rank #132 of 2245 rated models, from 76 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Nejumi 4 - JTruthfulQA | 77.62 | Score (%) | 99.4 |
| Nejumi 4 - jaster (2-shot) - NIILC | 62.84 | Character F1 (x100) | 99.4 |
| Nejumi 4 - jaster (0-shot) - JaMP | 83 | Exact match (%) | 98.7 |
| Nejumi 4 - Toxicity - Prohibited Acts | 98.44 | Criteria met (%) | 96.8 |
| Nejumi 4 - jaster (0-shot) - JCommonsenseQA | 100 | Exact match (%) | 96.1 |
| Nejumi 4 - jaster (2-shot) - JCoLA (out-of-domain) | 90 | Exact match (%) | 95.5 |
| Nejumi 4 - jaster (0-shot) - AIO | 93.83 | Character F1 (x100) | 93.5 |
| Nejumi 4 - ARC-AGI-2 | 82 | Accuracy (%) | 93.2 |
| Nejumi 4 - jaster (0-shot) - NIILC | 59.38 | Character F1 (x100) | 92.9 |
| Nejumi 4 - ARC-AGI-1 | 98 | Accuracy (%) | 92.2 |
| Nejumi 4 - jaster (0-shot) - JCoLA (out-of-domain) | 90 | Exact match (%) | 92.2 |
| SecureBio BioTIER Permit | 99.9 | Benign prompts answered (%) | 90.8 |
Interactive version: theaggregate.ai/model?slug=qwen3-8-2-4t-a95b-xhigh · How It Works · Data refreshed daily, snapshot 2026-11-16.