Qwen 3.5 35B A3B: benchmark results
Alibaba's open Qwen 3.5 35B A3B MoE (3B active). Provider: Alibaba. Released 2026-02-24. Access: Open.
Unified ELO 1600 ± 1, rank #265 of 1392 rated models, from 287 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (AndroidWorld_SR) | 71.1 | Score (%) | 100 |
| Roboflow Vision Evals - Visual Understanding | 79.1 | Pass Rate (%) | 98.7 |
| JudgeBench Reasoning | 96.94 | Accuracy (%) | 98 |
| MERA - RWSD | 82.31 | Accuracy (%) | 93.3 |
| MERA - ruModAr | 99.67 | EM (%) | 93.3 |
| MERA - ruMMLU | 86.68 | Accuracy (%) | 92.8 |
| MERA - MaMuRAMu | 88.32 | Accuracy (%) | 92.3 |
| MERA - ruHateSpeech | 89.81 | Accuracy (%) | 92.3 |
| LIBRA - MatreshkaNames * | 68.97 | Dataset Total Score (%) | 91.7 |
| LIBRA - ruBABILongQA1 | 68.38 | Dataset Total Score (%) | 91.7 |
| LIBRA - ruBABILongQA2 | 54.97 | Dataset Total Score (%) | 91.7 |
| LIBRA - ruBABILongQA3 * | 32.6 | Dataset Total Score (%) | 91.7 |
Interactive version: theaggregate.ai/model?slug=qwen-3-5-35b-a3b · How It Works · Data refreshed daily, snapshot 2026-09-05.