Qwen 2.5 Plus: benchmark results
Alibaba's proprietary Qwen2.5-generation API tier, offered through Model Studio alongside the open Qwen2.5 release. Provider: Alibaba. Released 2024-09-19. Access: API.
Aggregate Rating 1571 ± 27, rank #1396 of 10916 rated models, from 31 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BenchTable - Tech | 63.1 | Weighted Score (%) | 62.5 |
| BenchTable - STEM | 55.8 | Weighted Score (%) | 54.5 |
| BenchTable | 48.1 | Total Score (%) | 49.3 |
| BenchTable - Reasoning | 38.3 | Weighted Score (%) | 46.5 |
| AI Chess Leaderboard (Reasoning) | 649 | Elo | 41.4 |
| Chatbot Arena (Text - Hard Prompts (English)) | 1342 | Arena Score | 38.7 |
| Chatbot Arena (Text - Math) | 1303 | Arena Score | 38.6 |
| Chatbot Arena (Text - Mathematical) | 1315 | Arena Score | 38.5 |
| Chatbot Arena (Text - Life, Physical & Social Science) | 1337 | Arena Score | 38.3 |
| Chatbot Arena (Text - Business, Management & Financial Ops) | 1315 | Arena Score | 37.7 |
| Chatbot Arena (Text - Hard Prompts) | 1326 | Arena Score | 37.3 |
| Chatbot Arena (Text - Multi-Turn) | 1307 | Arena Score | 36.6 |
Interactive version: theaggregate.ai/model?slug=qwen-2-5-plus · How It Works · Data refreshed daily, snapshot 2026-11-08.