DeepSeek V4 Flash (Max) — benchmark results
DeepSeek V4 Flash evaluated at the max reasoning-effort setting. Provider: DeepSeek. Released 2026-04-23. Access: Open.
Unified ELO 1871 ± 23, rank #87 of 1776 rated models, from 36 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (CodeForces) | 100 | Score | 96.7 |
| CritPt | 7.1 | Accuracy (self-reported) | 91.7 |
| MathArena - APEX Shortlist 2025 | 89.36 | Accuracy (%) | 88.9 |
| ZeroEval GPQA Diamond | 88.1 | GPQA Diamond Score | 86.5 |
| LLM Stats (MathArena Apex) | 85.7 | Score (%) | 83.3 |
| AA-LCR | 63 | Score (self-reported) | 80.3 |
| LLM Stats (HMMT Feb 26) | 94.8 | Score (%) | 80 |
| MathArena - ARXIV_FALSE March | 19.64 | Accuracy (%) | 78.6 |
| LLM2014 Logic 2026-05 | 50.97 | Median Score | 76.3 |
| OckBench | 83 | Accuracy (%) | 76 |
| MathArena - APEX 2025 | 27.08 | Accuracy (%) | 75 |
| LLM2014 Logic 2026-04 | 56.54 | Median Score | 72.5 |
Interactive version: theaggregate.ai/model?slug=deepseek-v4-flash-max · How the rankings work · Data refreshed daily, snapshot 2026-07-22.