DeepSeek V4.1 Flash (1017) (Max): benchmark results
Provider: DeepSeek. Released 2026-10-17. Access: Open.
Aggregate Rating 1853 ± 19, rank #64 of 2168 rated models, from 62 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AI Coding Daily (OpenCode) - Total | 60.97 | Total points (max 70) | 100 |
| AA GDPval | 1745.77 | ELO | 98 |
| AA-Briefcase - Presentation Elo | 1602.45 | Elo | 97.8 |
| AutomationBench-AA | 66.64 | Guardrail-Adjusted Objective Completion (%) | 95.1 |
| Artificial Analysis Intelligence Index | 45.1 | Intelligence Index | 95 |
| AA Long Context Reasoning | 82.67 | Accuracy (%) | 93.8 |
| AutomationBench-AA - Strict | 36.68 | Strict Task Success (%) | 93.8 |
| AA CritPt | 24 | Accuracy (%) | 91.8 |
| AutomationBench-AA - Finance - Raw Objective Completion | 87.94 | Raw Objective Completion (%) | 91.6 |
| AA Humanity's Last Exam | 43.33 | Accuracy (%) | 91.5 |
| AA-Briefcase - Analytical Quality Elo | 1765.78 | Elo | 89.2 |
| AA-Omniscience Index - Software Engineering (SWE) - Rust | 68 | Omniscience Index | 88.7 |
Interactive version: theaggregate.ai/model?slug=deepseek-v4-1-flash-1017-max · How It Works · Data refreshed daily, snapshot 2026-10-20.