DeepSeek V3.2 Chat: benchmark results
Provider: DeepSeek. Released 2025-12-01. Access: Open.
Unified ELO 1741 ± 30, rank #256 of 2656 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MCPMark | 29.72 | Pass@1 (%) | 68.4 |
| AGI-Eval Community - Subject Knowledge | 85.33 | Accuracy (%) | 67.9 |
| AGI-Eval Community - General Reasoning | 96.82 | Accuracy (%) | 65.7 |
| AGI-Eval Community - Subject Reasoning (Chinese) | 85.29 | Accuracy (%) | 60.5 |
| AGI-Eval Community - Subject Reasoning (English) | 83 | Accuracy (%) | 58 |
| AGI-Eval Community - Subject Reasoning | 83.7 | Accuracy (%) | 57.5 |
| AGI-Eval Community - Cognition (Chinese) | 76.52 | Accuracy (%) | 53.6 |
| AGI-Eval Community - Learning (Chinese) | 78.16 | Accuracy (%) | 52.9 |
| AGI-Eval Community - Interaction (English) | 90.06 | Accuracy (%) | 50 |
| AGI-Eval Community - Interaction (Chinese) | 74.47 | Accuracy (%) | 44.9 |
| AGI-Eval Community - Objective Accuracy (Chinese) | 57.38 | Accuracy (%) | 44.2 |
| AGI-Eval Community - Objective Accuracy | 63.16 | Accuracy (%) | 42.9 |
Interactive version: theaggregate.ai/model?slug=deepseek-v3-2-chat · How It Works · Data refreshed daily, snapshot 2026-09-19.