Qwen 3.8 Flash: benchmark results
Provider: Alibaba. Released 2026-08-26. Access: API.
Unified ELO 1731 ± 1, rank #25 of 1392 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RealWorldQA | 88.5 | RealWorldQA (self-reported) | 99.3 |
| LLM Stats (MathVision) | 95.7 | Score (%) | 95.6 |
| LLM Stats Score | 49.19 | LLM Stats Score (conservative rating) | 94.6 |
| LLM Stats (ERQA) | 72.3 | Score (%) | 94 |
| AI Chess Leaderboard (Reasoning) | 1528 | Elo | 93.8 |
| LLM Stats (CharXiv-R) | 90.6 | Score (%) | 93.5 |
| ZeroEval GPQA Diamond | 91.7 | GPQA Diamond Score | 92.3 |
| LLM Stats (ClawEval-MM) | 60.4 | Score (%) | 90 |
| LLM Stats (Toolathlon) | 73.5 | Score (%) | 88.8 |
| LVBench | 76.6 | Score (self-reported) | 84.9 |
| ComplexConstraints | 43.3 | Score (%) | 81.7 |
| LLM Stats (Agents' Last Exam) | 51.2 | Score (%) | 76.7 |
Interactive version: theaggregate.ai/model?slug=qwen-3-8-flash · How It Works · Data refreshed daily, snapshot 2026-09-05.