Qwen 3.5 27B (Reasoning): benchmark results
Provider: Alibaba. Released 2026-02-24. Access: Open.
Unified ELO 1639 ± 1, rank #423 of 3363 rated models, from 189 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SEA-HELM (Burmese) - SEA-Safeguard | 89.24 | Normalized Score | 100 |
| SEA-HELM (Tamil) - SEA-IFEval | 95.7 | Normalized Score | 100 |
| SEA-HELM (Burmese) - SEA-IFEval | 85.71 | Normalized Score | 98.2 |
| SEA-HELM (Indonesian) - Natural Language Inference | 83.23 | Normalized Score | 98.2 |
| SEA-HELM (Tamil) - NLR | 75.31 | Normalized Score | 98.2 |
| SEA-HELM (Thai) - NLU | 64.55 | Normalized Score | 97.4 |
| SEA-HELM (English) - Knowledge | 75.65 | Normalized Score | 96.5 |
| SEA-HELM (English) - SuperGPQA | 60.01 | Normalized Score | 96.5 |
| SEA-HELM (Indonesian) - Global MMLU Lite | 85.23 | Normalized Score | 96.5 |
| SEA-HELM (Indonesian) - SEA-MT-Bench (LLM Judge) | 87.29 | Normalized Score | 96.5 |
| SEA-HELM (Tamil) - Causal Reasoning | 95.49 | Normalized Score | 96.5 |
| SEA-HELM (Tamil) - Natural Language Inference | 55.14 | Normalized Score | 96.5 |
Interactive version: theaggregate.ai/model?slug=qwen-3-5-27b-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-23.