Qwen 2.5 72B: benchmark results
Alibaba's open 72B Qwen2.5 flagship (September 2024) with a 128K context and much stronger coding and math than Qwen2. Provider: Alibaba. Released 2024-09-19. Access: Open.
Aggregate Rating 1582 ± 7, rank #499 of 1656 rated models, from 398 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| NeedleBench | 81.02 | Overall 128K score (imported) | 100 |
| Open PL LLM - PSC (multiple choice, 5-shot) | 97.25 | Binary F1 (%) | 100 |
| RSPC - Temporal Phase Classification (Zero-Shot) | 35.1 | Macro-F1 (%) on classification of the relationship's tempora | 100 |
| StressEval - Knowledge Graph | 42.4 | Exact-match accuracy (%) on the knowledge-graph reasoning sp | 100 |
| StressEval - Knowledge Graph Knowledge Stress | 33.9 | Exact-match accuracy (%) on the knowledge-graph knowledge-st | 100 |
| SurgiQ | 68.08 | Micro accuracy over all 13,055 questions: accuracy (%) on Su | 100 |
| SurgiQ - Case-Based | 67.9 | Case-based questions (6,156): accuracy (%) on SurgiQ's four- | 100 |
| SurgiQ - Critical Care and Emergency | 68.14 | Critical care and emergency surgery questions (590): accurac | 100 |
| SurgiQ - Negative | 68.31 | Negative (all-except) questions (2,153): accuracy (%) on Sur | 100 |
| SurgiQ - Neurosurgery | 68.51 | Neurosurgery questions (2,204): accuracy (%) on SurgiQ's fou | 100 |
| SurgiQ - Orthopedic Surgery | 62.99 | Orthopedic surgery questions (1,189): accuracy (%) on SurgiQ | 100 |
| SurgiQ - Robotic Surgery | 69.08 | Robotic surgery questions (1,643): accuracy (%) on SurgiQ's | 100 |
Interactive version: theaggregate.ai/model?slug=qwen-2-5-72b · How It Works · Data refreshed daily, snapshot 2026-10-20.