# Tencent Hy4 - BenchAudit True Index Profile

**Integrity Profile: Clean Generalization**

> The Verdict: Performs Better on Fresh Private Evals Than Legacy Tests. Tencent Hy4 (evaluated at default effort) outperforms its legacy benchmark standing (53.1p) when moved to private, leak-resistant evaluations (52.1p), posting a -4.4p Low-Tier Boost across 3 matched domains, particularly in Math & Scientific Reasoning (-30.6p boost)—a strong indicator of genuine generalization.

- **Canonical URL:** https://benchaudit.org/models/tencent-hy4/
- **True Index Rank:** #22 of 44
- **Creator:** Tencent
- **Weights:** Open Weights
- **Clean Evals Avg (Tiers S & A):** 52.1p (7 evals)
- **Legacy Benchmarks Avg (Tiers C & D):** 53.1p (4 evals)
- **Low-Tier Boost (Matched Delta):** -4.4p
- **Updated:** 2026-11-22

## True Index: Domain Rankings

| Capability Domain | Sub-Rank | Clean Evals (S/A) Avg | Top-Tier Evals |
| --- | ---: | ---: | ---: |
| Agentic & Enterprise | #28 | 47.0p | 2 |
| Coding & SWE | #30 | 43.7p | 3 |
| Math & Scientific Reasoning | #7 | 81.8p | 1 |

## 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 | 2 (47.0p) | 1 (51.0p) | +4.0p |
| Coding & SWE | 3 (43.7p) | 1 (57.1p) | +13.4p |
| Math & Scientific Reasoning | 1 (81.8p) | 2 (51.2p) | -30.6p |

## Audited Benchmark Performance of Tencent Hy4

| Benchmark | Benchmark Integrity Status | Domain | Rank | Percentile |
| --- | --- | --- | ---: | ---: |
| [MLCR-AA](https://benchaudit.org/benchmarks/mlcr-aa/) | Private / Fresh | Math & Scientific Reasoning | #10/46 | 81.8p |
| [FrontierCode-AA 2.0](https://benchaudit.org/benchmarks/frontiercode-aa-2-0/) | Private / Fresh | Coding & SWE | #16/46 | 68.4p |
| [AutomationBench-AA](https://benchaudit.org/benchmarks/automationbench-aa/) | Private / Fresh | Agentic & Enterprise | #21/50 | 59.2p |
| [HLE-Diamond](https://benchaudit.org/benchmarks/hle-diamond/) | Semi-Exposed | General Knowledge & QA | #21/49 | 58.3p |
| [Terminal-Bench 4.0](https://benchaudit.org/benchmarks/terminal-bench-4-0/) | Public / Aging | Coding & SWE | #22/50 | 57.1p |
| [Terminal-Bench-Science 0.1](https://benchaudit.org/benchmarks/terminal-bench-science-0-1/) | Public / Aging | Math & Scientific Reasoning | #21/41 | 53.6p |
| [GDP.pdf](https://benchaudit.org/benchmarks/gdp-pdf/) | Public / Aging | Agentic & Enterprise | #25/50 | 51.0p |
| [BioMysteryBench](https://benchaudit.org/benchmarks/biomysterybench/) | Public / Aging | Math & Scientific Reasoning | #14/23 | 48.8p |
| [CyberBench v1.1](https://benchaudit.org/benchmarks/cyberbench/) | Private / Fresh | Coding & SWE | #18/25 | 39.0p |
| [AA-Omniscience v1.1](https://benchaudit.org/benchmarks/aa-omniscience-v1-1/) | Private / Fresh | General Knowledge & QA | #31/47 | 36.0p |
| [AA-Briefcase v1.1](https://benchaudit.org/benchmarks/aa-briefcase-v1-1/) | Private / Fresh | Agentic & Enterprise | #33/50 | 34.7p |
| [Code Migration v1.1](https://benchaudit.org/benchmarks/code-migration-v1-1/) | Private / Fresh | Coding & SWE | #19/24 | 23.8p |
