Hy4

Hy4 review navigator

Official benchmarks, independent analysis, and community reports about Hy4, clearly separated from Tabbit's own testing.

12 source-checked resourcesOfficial · Media · Community

Official

1 source-checked resources

Media

5 source-checked resources
MediaMercor APEX

Mercor APEX-Agents Professional Task Benchmark: Hy4 High Score Snapshot

As of 2026-11-16, Mercor's APEX-Agents leaderboard lists Hy4 High with a Pass@1 of 77.6% ±4.5% and a Mean Score of 85.1% ±3.6%. This is a leaderboard snapshot of Hy4 on professional, multi-step agent tasks in APEX-Agents 1.1; it does not represent general-purp.

MediaVals AI

Vals AI: Hy4's Cross-domain Results on Vals Index

On the 2026-11-15 leaderboard, Vals AI ranked Tencent Hy4 (the production release launched on November 4) at 59.35%, placing it 17/38 overall. Its performance varied across coding, finance, financial modeling, and tax benchmarks. This is a result for a specifi.

MediaPlotly official blog

Plotly: Tencent Hunyuan Hy4 Data Analytics Benchmark

Plotly reports that Hy4 answered 42/43 questions (98%) correctly on its data analytics benchmark, tying Grok 4.7 and Kimi K3.1 Pro by overall score; only GPT-6.1 Sol and GPT-6 Astra scored higher at 43/43. Hy4 completed the full benchmark for $2.11, the lowest.

MediaTencent Cloud Developer Community

Hands-on Review of Hy4 on Tencent Cloud Developer Community: Multitask Performance and Evidence Limits

This hands-on WorkBuddy report covers coding, visual reasoning, planning, and design, and records several task inputs, durations, and specific errors. However, it omits model and client configurations, complete execution traces, and a consistent scoring method.

MediaTencent Cloud Developer Community

Tencent Cloud Developer Community: Hands-on Trials of Hy4 GA for Image and 3D Generation

The author describes hands-on experience with Hy4 GA across image generation and two types of 3D scenarios: it produced a relatively complete industrial plant, but slowly and with mediocre detail and aesthetics; its sandbox game had a distinctive visual style.

Community

6 source-checked resources
CommunityArena.ai official Agent Arena leaderboard and X

Arena Agent Arena Real-World Session Evaluation: Hy4 Delivers a +5.4% Net Lift at a Median Cost of $0.57 per Task

Arena said on November 6 that Hy4 achieved a +5.4% net lift across more than 12K real-user Agent sessions in its evaluation, at a median cost of $0.57 per task. These figures reflect the real tasks and runtime environment covered by Agent Arena; they are not a.

CommunityX; Artificial Analysis model and evaluation methodology pages

Artificial Analysis: Hy4 Intelligence Index v5 and Evaluation Costs

On November 5, Artificial Analysis reported that Hy4 scored 35 on Intelligence Index v5, alongside results for agentic coding, knowledge question answering, and evaluation costs. The post did not specify a more precise v5 subversion. As of November 16, the met.

CommunityX long-form post

Three Models, the Same Tasks: A Hands-on Comparison of Hy4, DeepSeek V4.1 Flash, and Qwen4

The author reports 18 same-task tests: Hy4 led on small projects, some Chinese writing, and close-reading tasks; DeepSeek V4.1 Flash led on design and promotional copy scores; and Qwen4 led in correct answers on mazes, grid route planning, and research-level m.

CommunityX

Marcel: Hy4 vs. GPT 6.1 Terra on a Single-File 3D Castle

The author says Hy4 (max) completed the task in about 6 minutes at an estimated cost of about $0.03, while GPT 6.1 Terra (max) took about 16 minutes and cost about $0.44. The author considers Hy4 faster and cheaper, but says its 3D coding quality still trails .

CommunityX

Leo: Hy4 vs. DeepSeek v4.1 Flash in a Browser Sandbox Game

The author says that, given the same prompt to generate a browser-based Minecraft-style sandbox game, the Hy4 version could not detect the mouse and had reversed directional controls. He considers this result clearly behind DeepSeek v4.1 Flash. The original po.

CommunityX

Xueyu: Hy4 GA vs. Preview on the Same Pelican Animation Prompt

The author says he used the same prompt for Hy4 GA and an August 31 run of Hy4 Preview. He found the GA version's pedaling motion and foot placement more convincing, but flagged problems with the beak and sunglasses. The post lists a 19-second video. We could .

Hy4

Use and compare models in Tabbit

Official benchmarks, independent analysis, and community reports about Hy4, clearly separated from Tabbit's own testing.