# DeepSeek V4 Pro 0813 - Benchmaxxing Radar Profile

**Integrity Profile: Clean Generalization**

> The Verdict: Performs Better on Fresh Private Evals Than Legacy Tests. DeepSeek V4 Pro 0813 (evaluated at max effort) outperforms its legacy benchmark standing (31.5p) when moved to private, leak-resistant evaluations (36.0p), posting a -9.2p Low-Tier Boost across 3 matched domains, particularly in Agentic & Enterprise (-29.3p boost) and Coding & SWE (-20.6p boost)—a strong indicator of genuine generalization.

- **Canonical URL:** https://benchaudit.org/models/deepseek-v4-pro__max/
- **Benchmaxxing Radar Rank (100-day cohort):** #43 of 47
- **Creator:** DeepSeek
- **Weights:** Open Weights
- **Clean Evals Avg (Tiers S & A):** 36.0p (10 evals)
- **Legacy Benchmarks Avg (Tiers C & D):** 31.5p (4 evals)
- **Low-Tier Boost (Matched Delta):** -9.2p
- **Updated:** 2026-11-22

## Head-to-Head: Clean Evals vs. Legacy Benchmarks

| Capability Domain | Clean Evals (Tiers S & A) | Legacy Benchmarks (Tiers C & D) | Public Lift (Boost) |
| --- | ---: | ---: | ---: |
| Agentic & Enterprise | 3 (46.3p) | 2 (17.0p) | -29.3p |
| Coding & SWE | 3 (45.1p) | 1 (24.5p) | -20.6p |
| Math & Scientific Reasoning | 2 (30.6p) | 1 (53.1p) | +22.5p |

## Audited Benchmark Performance of DeepSeek V4 Pro 0813

| Benchmark | Benchmark Integrity Status | Domain | Rank | Percentile |
| --- | --- | --- | ---: | ---: |
| [EnterpriseOps-Gym-AA](https://benchaudit.org/benchmarks/enterpriseops-gym-aa/) | Private / Fresh | Agentic & Enterprise | #1/13 | 100.0p |
| [CyberBench v1.1](https://benchaudit.org/benchmarks/cyberbench/) | Private / Fresh | Coding & SWE | #9/25 | 68.2p |
| [GPQA Diamond](https://benchaudit.org/benchmarks/gpqa-diamond/) | Public / Aging | Math & Scientific Reasoning | #8/17 | 53.1p |
| [Code Migration v1.1](https://benchaudit.org/benchmarks/code-migration-v1-1/) | Private / Fresh | Coding & SWE | #14/24 | 43.8p |
| [CritPt v2](https://benchaudit.org/benchmarks/critpt-v2/) | Private / Fresh | Math & Scientific Reasoning | #25/40 | 38.5p |
| [HLE-Diamond](https://benchaudit.org/benchmarks/hle-diamond/) | Semi-Exposed | General Knowledge & QA | #35/49 | 29.2p |
| [AutomationBench-AA](https://benchaudit.org/benchmarks/automationbench-aa/) | Private / Fresh | Agentic & Enterprise | #36/50 | 28.6p |
| [ARC-AGI-2](https://benchaudit.org/benchmarks/arc-agi-2/) | Private / Fresh | General Knowledge & QA | #18/22 | 28.2p |
| [Terminal-Bench 4.0](https://benchaudit.org/benchmarks/terminal-bench-4-0/) | Public / Aging | Coding & SWE | #38/50 | 24.5p |
| [GDPval-AA v2.1](https://benchaudit.org/benchmarks/gdpval-aa-v2-1/) | Public / Aging | Agentic & Enterprise | #33/43 | 23.8p |
| [FrontierCode-AA 2.0](https://benchaudit.org/benchmarks/frontiercode-aa-2-0/) | Private / Fresh | Coding & SWE | #36/46 | 23.4p |
| [FrontierMath](https://benchaudit.org/benchmarks/frontiermath/) | Private / Fresh | Math & Scientific Reasoning | #24/27 | 22.8p |
| [AA-Omniscience v1.1](https://benchaudit.org/benchmarks/aa-omniscience-v1-1/) | Private / Fresh | General Knowledge & QA | #40/47 | 15.3p |
| [AA-Briefcase v1.1](https://benchaudit.org/benchmarks/aa-briefcase-v1-1/) | Private / Fresh | Agentic & Enterprise | #45/50 | 10.2p |
| [GDP.pdf](https://benchaudit.org/benchmarks/gdp-pdf/) | Public / Aging | Agentic & Enterprise | #45/50 | 10.2p |
