Qwen 3.8 2.4T A95B (xHigh): benchmark results
Provider: Alibaba. Released 2026-08-12. Access: Open.
Aggregate Rating 1826 ± 20, rank #94 of 2163 rated models, from 72 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Nejumi 4 - JTruthfulQA | 77.62 | Score (%) | 99.3 |
| Nejumi 4 - jaster (2-shot) - NIILC | 62.84 | Character F1 (x100) | 99.3 |
| Nejumi 4 - jaster (0-shot) - JaMP | 83 | Exact match (%) | 98.6 |
| Nejumi 4 - jaster (2-shot) - JCoLA (out-of-domain) | 90 | Exact match (%) | 98.2 |
| Nejumi 4 - ARC-AGI-2 | 82 | Accuracy (%) | 96.5 |
| Nejumi 4 - Toxicity - Prohibited Acts | 98.44 | Criteria met (%) | 96.5 |
| Nejumi 4 - jaster (0-shot) - JCommonsenseQA | 100 | Exact match (%) | 96.1 |
| Nejumi 4 - jaster (0-shot) - JCoLA (out-of-domain) | 90 | Exact match (%) | 94.7 |
| Nejumi 4 - jaster (0-shot) - NIILC | 59.38 | Character F1 (x100) | 94.4 |
| Nejumi 4 - HLE (Japanese) - Calibration Error | 44 | Calibration error (%) | 94 |
| Nejumi 4 - ARC-AGI-1 | 98 | Accuracy (%) | 93.3 |
| Nejumi 4 - jaster (0-shot) - AIO | 93.83 | Character F1 (x100) | 93 |
Interactive version: theaggregate.ai/model?slug=qwen-3-8-2-4t-a95b-xhigh · How It Works · Data refreshed daily, snapshot 2026-10-19.