Qwen 3 VL 8B (Thinking) — benchmark results
Alibaba Qwen 3 VL 8B vision-language model evaluated with thinking enabled. Provider: Alibaba. Released 2025-07-01. Access: Open.
Unified ELO 1543 ± 13, rank #644 of 1776 rated models, from 97 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OCR-Robust | 79.67 | OCR1.0 Clean (self-reported) | 93.3 |
| MedLayXPlain | 64.6 | S (self-reported) | 83.9 |
| MathVision | 59.6 | Overall Accuracy (%) | 73.9 |
| CFMME | 53.85 | Average (self-reported) | 73.3 |
| K-MetBench | 71.7 | Accuracy (self-reported) | 72.4 |
| LLM Stats (MuirBench) | 76.8 | Score (%) | 70 |
| LLM Stats (WritingBench) | 85.5 | Score (%) | 67.9 |
| LLM Stats (BLINK) | 68.7 | Score (%) | 66.7 |
| AA Omniscience - Software Engineering (SWE) - Go | 24 | Accuracy (%) | 66.1 |
| AA Omniscience - Software Engineering (SWE) - PHP | 32 | Accuracy (%) | 65 |
| CritPt | 0.3 | Accuracy (self-reported) | 65 |
| AA Omniscience - Software Engineering (SWE) - Swift | 44 | Accuracy (%) | 63.3 |
Interactive version: theaggregate.ai/model?slug=qwen-3-vl-8b-thinking · How the rankings work · Data refreshed daily, snapshot 2026-07-22.