Qwen 3 Max (Preview) (Thinking): benchmark results

Preview thinking snapshot of Qwen 3 Max. Provider: Alibaba. Released 2026-01-27. Access: API.

Unified ELO 1631 ± 1, rank #440 of 3078 rated models, from 99 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA Omniscience - Software Engineering (SWE) - R30Accuracy (%)88.7
AA Omniscience - Software Engineering (SWE) - C56Accuracy (%)87.5
AA Omniscience - Software Engineering (SWE) - PHP42Accuracy (%)87.2
AA Omniscience - Software Engineering (SWE) - TypeScript42.22Accuracy (%)86.9
AA Omniscience - Software Engineering (SWE) - Kotlin32Accuracy (%)86.5
AA Omniscience - Software Engineering (SWE) - Python34Accuracy (%)85.7
AA Omniscience - Software Engineering (SWE) - Dart30Accuracy (%)85.2
AA Omniscience - Software Engineering (SWE) - Julia28Accuracy (%)85.2
AA Omniscience - Software Engineering (SWE) - JavaScript41.82Accuracy (%)85
AA Omniscience - Software Engineering (SWE) - Java25Accuracy (%)84.6
AA MMLU-Pro82.45Accuracy (%)80.5
AGI-Eval Community - Interaction (Chinese)83.23Accuracy (%)79.3

Interactive version: theaggregate.ai/model?slug=qwen-3-max-preview-thinking · How It Works · Data refreshed daily, snapshot 2026-09-19.