DeepSeek V3.2 (Non-reasoning) — benchmark results
DeepSeek V3.2 evaluated with reasoning disabled. Provider: DeepSeek. Released 2025-12-01. Access: Open.
Unified ELO 1619 ± 25, rank #436 of 1776 rated models, from 80 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI Leaderboard | 55.75 | UGI Score | 96.6 |
| UGI - Writing | 56.3 | Writing Score | 91.7 |
| UGI - Natural Intelligence | 47.85 | NatInt Score | 90.1 |
| AA MMLU-Pro | 83.71 | Accuracy (%) | 86.3 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 34 | Accuracy (%) | 78.2 |
| AA Terminal-Bench Hard | 32.58 | Accuracy (%) | 76.6 |
| AA Omniscience - Software Engineering (SWE) - Dart | 32 | Accuracy (%) | 75.8 |
| AA Omniscience - Software Engineering (SWE) - Go | 30 | Accuracy (%) | 74.3 |
| UGI - Willingness (W/10) | 7.2 | W/10 Score | 73.6 |
| AA Omniscience - Software Engineering (SWE) - C | 54 | Accuracy (%) | 73.5 |
| AA Omniscience - Software Engineering (SWE) - R | 24 | Accuracy (%) | 72 |
| AA Omniscience - Humanities & Social Sciences | 26.5 | Accuracy (%) | 71.7 |
Interactive version: theaggregate.ai/model?slug=deepseek-v3-2-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.