Skip to content
Model Index
  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder
LabsMethodology
  1. Models
  2. Tencent

Hy4 preview

Tencent · released Aug 28, 2026

FlagshipOpen weightsPreview

Tencent's newest and largest open-weight model, with a 1M-token context, aimed at coding agents, office and analysis work and research tasks.

Best for: Apache-licensed open weights · Coding agents · Long-context text work

Documentation
Intelligence Index
—
Not yet evaluated
Price per 1M tokens
$0.83 in $2.5 out
Blended $1.25 at 3:1
Context window
1M
Up to 64K output
Output speed
—
Not measured

Capability profile

One 0–100 score per area, placing this model between the weakest and strongest tracked model across that area's benchmarks.

  • OverallNot enough data

    Based on 1 of 3 benchmarks

  • Coding69 · #7 of 39

    Based on 3 of 5 benchmarks

  • Agents & knowledge work63 · #20 of 40

    Based on 5 of 6 benchmarks

  • Reasoning & scienceNot enough data

    Based on 0 of 2 benchmarks

  • Knowledge & reliabilityNot enough data

    Based on 0 of 2 benchmarks

  • VisionNot enough data

    Based on 0 of 2 benchmarks

Benchmark results

All third-party results on record. Rank is among the tracked models that have been tested on that benchmark.

Overall

  • Artificial Analysis Intelligence IndexNot tested
  • Vals Index49.9%

    #18 of 26· $1.44 per test· Vals AI

  • Arena · TextNot tested

Coding

  • Terminal-Bench 4.0Not tested
  • SciCodeNot tested
  • Arena · WebDev1632 ±10

    #12 of 34· 4,798 votes· Arena

  • Vibe Code Bench77.5%

    #19 of 26· Vals AI

  • Code Migration46.8%

    #10 of 26· Vals AI

Agents & knowledge work

  • GDPval-AANot tested
  • Arena · Agent+4.0 ±0.8

    #8 of 29· Arena

  • Finance Agent v255.1%

    #10 of 26· Vals AI

  • Excel Modeling Benchmark58.1%

    #17 of 26· Vals AI

  • Legal Research Bench45.2%

    #11 of 26· Vals AI

  • Tax Agent Bench63.7%

    #13 of 26· Vals AI

Reasoning & science

  • Humanity's Last ExamNot tested
  • CritPtNot tested
  • ARC-AGI-2saturatingNot tested
  • ARC-AGI-3Not tested
  • GPQA DiamondsaturatingNot tested

Knowledge & reliability

  • AA-OmniscienceNot tested
  • AA-LCRNot tested

Vision

  • MMMU-ProNot tested
  • Arena · VisionNot tested

Specifications

Lab
Tencent
Released
Aug 28, 2026
Availability
Preview
Context window
1,000,000 tokens
Max output
64,000 tokens
Accepts
Text
Knowledge cutoff
Not published
Weights
Open
Licence
Apache 2.0
Parameters
770B total · 49B active
Reasoning control
Effort high or off
API model ID
hy4-preview

Tencent labels this an early version with known issues, such as reasoning for longer than necessary.

Pricing

Input / 1M tokens
$0.83
Cached input / 1M
$0.04
Output / 1M tokens
$2.5
Blended (3:1)
$1.25

Worth knowing

  • Release date is the first public listing; Tencent published no dated announcement.

Official sources

  • Documentation
  • Weights

23 benchmarks tracked. See methodology and sources.

Model Index

An independent snapshot of frontier SI models and how they perform on third-party benchmarks.

Data as of Oct 8, 2026

Explore

  • All models
  • Compare
  • Benchmarks
  • Model finder
  • Labs
  • Methodology & sources

Sources

Benchmark results come from Artificial Analysis, Arena, Vals AI and ARC Prize. Specifications and prices come from each lab's own documentation.

Not affiliated with any lab or evaluator. Names and logos are trademarks of their respective owners and are used only to identify them. Scores change often — check the linked source before making a decision that depends on one.

  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder
  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder
  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder