DeepSeek V4.1 Flash (Non-reasoning): benchmark results
Provider: DeepSeek. Released 2026-09-10. Access: Open.
Unified ELO 1623 ± 1, rank #341 of 1935 rated models, from 63 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA-Omniscience Index - Software Engineering (SWE) - Go | 30 | Omniscience Index | 81.7 |
| AA-Omniscience Hallucination Rate | 53.99 | Hallucination Rate (%) | 78.6 |
| ParseBench - Semantic Formatting | 66.05 | Semantic Formatting Score | 77.9 |
| Artificial Analysis Intelligence Index | 24.67 | Intelligence Index | 77.3 |
| AA-Omniscience Index - Business | -6 | Omniscience Index | 76.6 |
| AA-Omniscience Index - Software Engineering (SWE) - Python | 13.5 | Omniscience Index | 75.4 |
| AA-Omniscience Index - Software Engineering (SWE) - JavaScript | 15.45 | Omniscience Index | 74.5 |
| AutomationBench-AA - Finance - Raw Objective Completion | 79.68 | Raw Objective Completion (%) | 73.7 |
| AA GDPval | 1327.69 | ELO | 72.8 |
| AA-Omniscience Index - Humanities & Social Sciences | -7.9 | Omniscience Index | 72.6 |
| AA-Omniscience Index - Software Engineering (SWE) - HTML | 20 | Omniscience Index | 72.3 |
| AA-Omniscience Index - Law | -12.2 | Omniscience Index | 71.7 |
Interactive version: theaggregate.ai/model?slug=deepseek-v4-1-flash-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-25.