‹ PublicAI Index
The LLM benchmark aggregator.
Hy4
Tencent · open weights
Strongest in Biology research (#16 of 31), weakest in Professional (#73 of 194). Above par in 9 of 11 scopes. Among the models it meets almost everywhere, it finishes behind Gemini 4 Argon and Muse Wonder Full and ahead of Grok 4.7 and Gemini 3.8 Flash.
§ 1 · Profile
What it is good at
Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.
Coding56.7−9.4#25/1791/6
Agentic coding57.2−9.5#23/1731/5
Reasoning52.8−12.6#68/1951/5
Mathematics54.3−9.5#63/1521/2
Professional51.9−10.5#73/1941/2
Biology research49.5−21.2#16/311/1
IT operations40.8−33#25/331/1
Cybersecurity50.4−13.5#31/471/1
Legal54.6−11.3#49/1731/1
Finance54.7−7.7#52/1621/1
Medical52.8−13.7#65/1491/1
§ 2 · Head to head
What it beats, and what beats it
The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Hy4, left for the other.
§ 3 · Sources
Where the numbers come from
1 publication, 12 figures. Every one links to the page it was read from.
Vals · Legal Research Bench 41.35%Vals · LegalBench 83.76%Vals · Harvey Legal Agent Benchmark 12.5%Vals · Finance Agent 55.06%Vals · MedCode 43.25%Vals · MedScribe 83.6%Vals · BioMysteryBench 69.26%Vals · CyberBench 65.36%Vals · SRE Bench 2.29%Vals · Vibe Code Bench 81.01% (OpenHands)Vals · Code Migration 47.43%Vals · ProofBench 75%
Badge
[](https://publicai.io/model-index/m/hy4)