Qwen 3.5 27B — benchmark results
Alibaba's dense open Qwen 3.5 27B multimodal model. Provider: Alibaba. Released 2026-02-24. Access: Open.
Unified ELO 1670 ± 13, rank #321 of 1776 rated models, from 207 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EgoCoT-Bench | 71.28 | Mean (self-reported) | 100 |
| LLM Stats (Hallusion Bench) | 70 | Score (%) | 100 |
| SVFSearch | 95.4 | Overall Acc. (self-reported) | 100 |
| VANTAGE-Bench - Spatial | 79.04 | Spatial (%) | 100 |
| AI for Education Pedagogy - Technology | 88.68 | Accuracy (%) | 98.9 |
| AI for Education Pedagogy - Science | 93.99 | Accuracy (%) | 98 |
| FMNB Leaderboard | 100 | Score | 97.9 |
| StemBind | 41.3 | F Overall (self-reported) | 95.7 |
| MATH-MC Level 3 | 99.37 | Accuracy (%) | 95.6 |
| LLM Stats (MathVista-Mini) | 87.8 | Score (%) | 95.5 |
| K-MetBench | 83 | Accuracy (self-reported) | 94.8 |
| MATH-MC Level 2 | 99.09 | Accuracy (%) | 94.1 |
Interactive version: theaggregate.ai/model?slug=qwen-3-5-27b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.