1 min read
Add as a preferred source on Google

Tencent’s Hy4 Matches Grok 4.7 at 35 on AI Intelligence Index

Moonshot AI’s Kimi K3.1 Pro scored 46, 11 points above Hy4, while Hy4 posted a 31% result on FrontierCode 2.0.

An anonymous hand tests a laptop on a stark evaluation bench / TokenPost.ai
An anonymous hand tests a laptop on a stark evaluation bench / TokenPost.ai

Tencent’s Hy4 scored 35 on the Artificial Analysis Intelligence Index v5, matching Grok 4.7 at its xhigh setting and ranking second among models from China-based labs.

Moonshot AI’s Kimi K3.1 Pro led that group with a score of 46, an 11-point gap over Hy4. The index combines results from multiple evaluations into a composite score.

Hy4 also recorded 31% on FrontierCode 2.0, an agentic coding benchmark, and scored 3.04 on a finance evaluation with 579 tasks.

Tencent released Hy4’s full version Nov. 4, after an Aug. 28 preview. The release added vision, tool calling and computer use, expanding on Hy4’s cross-application computer controls.

Tencent plans to make Hy4’s weights and system cards available, giving researchers materials to examine its design and behavior.

Simon Yoon

Reporter

Simon Yoon reports on blockchain technology for TokenPost. Send corrections or tips to info@tokenpost.com.

Loading…