Tess-v2.5.2-Qwen2-72B — benchmark results
Migel Tissera's Tess general-purpose fine-tune of Qwen2 72B, trained on synthetic data for reasoning, coding, and math. Provider: Other. Released 2024-06-13. Access: Open.
Unified ELO 1495 ± 44, rank #821 of 1776 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MMLU-Pro | 50.68 | Score | 98.7 |
| Open LLM Leaderboard - BBH | 52.31 | Score | 95.1 |
| Open LLM Leaderboard - GPQA | 13.42 | Score | 87.9 |
| Open LLM Leaderboard - MATH Level 5 | 29.38 | Score | 81.4 |
| Open LLM Leaderboard - MuSR | 10.89 | Score | 55.6 |
| Open LLM Leaderboard - IFEval | 44.94 | Score | 49.3 |
| BigCodeBench | 37 | Pass@1 (%) | 41.3 |
Interactive version: theaggregate.ai/model?slug=tess-v2-5-2-qwen2-72b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.