Hy3-preview
Tencent · 2 thinking configurations
Intelligence22.7
Input / 1M$0.06
Output / 1M$0.21
Context256K
Tariff: Tencent · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost2 measured configurations · 0 without both values
Selected: thinking thinking · Click a point to switch configuration
Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to...
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
22.7 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
86.7 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
27.8 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
92.7 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
63.1 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
27.9 %Specifications from the catalog
Inputtext
Outputtext
Tool callingYes
Structured outputNo
SpeedNot available
First tokenNot available
Maximum output235.9K
Catalog listing date2026-04-22