DeepSeek V3.1 (Non-reasoning) — benchmark results
DeepSeek V3.1 evaluated with reasoning disabled. Provider: DeepSeek. Released 2025-08-21. Access: Open.
Unified ELO 1627 ± 22, rank #419 of 1776 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI Leaderboard | 51.4 | UGI Score | 92.3 |
| UGI - Writing | 51.06 | Writing Score | 89.4 |
| UGI - Natural Intelligence | 45.29 | NatInt Score | 89 |
| AA MMLU-Pro | 83.28 | Accuracy (%) | 84 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 36 | Accuracy (%) | 80.5 |
| AA Omniscience - Software Engineering (SWE) - Go | 32 | Accuracy (%) | 76.8 |
| AA Omniscience - Software Engineering (SWE) - Java | 24 | Accuracy (%) | 71.2 |
| AA Omniscience - Humanities & Social Sciences | 26.2 | Accuracy (%) | 71 |
| AA Omniscience - Software Engineering (SWE) - Julia | 24 | Accuracy (%) | 70.4 |
| AA Omniscience - Software Engineering (SWE) - Python | 30.5 | Accuracy (%) | 70.4 |
| AA Omniscience - Software Engineering (SWE) - R | 22 | Accuracy (%) | 69.4 |
| AA Omniscience - Software Engineering (SWE) - C | 49 | Accuracy (%) | 68.8 |
Interactive version: theaggregate.ai/model?slug=deepseek-v3-1-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.