Tencent, open weights

Hy4 preview

Hy4 preview by Tencent ranks 73rd of 354 ranked models on the Noometry Index as of October 2026, with a score of 45.3. Its strongest category is coding, where it ranks 38th. API pricing starts at $0.75 per million input tokens and $2.25 per million output tokens, with a 1.05M-token context window.

Last verified

Specifications

Noometry rank
#73 of 354
Index score
45.3
Evidence
Reported 3 results
Provider
Tencent
Released
August 28, 2026
Weights
Open weights
Reasoning
Yes
Context window
1.05M
Max output
64K
Input price
$0.75 / M
Output price
$2.25 / M
Blended price
$1.13 / M
Output speed
Not measured
Value
#118 of 219
Knowledge cutoff
Unknown
Input
text

Category scores

Each category score combines every public result we have in that category.

Hy4 preview category scores
  1. Coding 51.6
  2. Reasoning 31.9
  3. Math 55.7
Hy4 preview category ranks
CategoryScoreRankResults
Coding51.6#381
Reasoning31.9#791
Math55.7#421

Strengths and weaknesses

Categories where Hy4 preview places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

Hy4 preview: strongest categories
CategoryScorevs medianRank
Coding51.6+12.9#38 of 340, top 12%

Weakest categories

Hy4 preview: weakest categories
CategoryScorevs medianRank
Reasoning31.9+8.3#79 of 350, top 23%

Closest competitors

The models ranked just above and below Hy4 preview. When scores are this close, price and speed are often the better way to choose.

Models ranked closest to Hy4 preview
ModelRankScoreBlended $/MSpeed
GPT-5.3 Codex#6945.8$4.81—Compare
Kimi K2 Thinking Turbo#7045.8——Compare
Qwen3.5 Max Preview#7145.3——Compare
Qwen3.7 Plus#7245.3$0.70—Compare
MiMo-V2.5-Pro#7445.2$0.54—Compare
Gemini 2.5 Pro#7545.0$3.445Compare
GPT-5.4 mini#7645.0$1.6910Compare
Amazon Nova Experimental Chat 26 02 10#7744.5——Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

Hy4 preview Coding benchmark results
BenchmarkScorePositionSettingSourceDate
LMArena WebDev1632#14 of 113, top 13%LMArena2026-10-08

Reasoning

Hy4 preview Reasoning benchmark results
BenchmarkScorePositionSettingSourceDate
NYT Connections (extended)68.2%#52 of 91, top 58%Lech Mazur benchmarks

Math

Hy4 preview Math benchmark results
BenchmarkScorePositionSettingSourceDate
ProofBench75%#13 of 77, top 17%Epoch AI

API pricing by provider

Hy4 preview API prices
RouteInput $/MOutput $/MCached input $/MChecked
deepinfra$0.83$2.50$0.0422026-10-10
openrouter$0.75$2.25$0.03782026-10-10

Compare Hy4 preview

Other Tencent models

Frequently asked questions

How good is Hy4 preview?

Hy4 preview by Tencent ranks 73rd of 354 ranked models on the Noometry Index as of October 2026, with a score of 45.3. Its strongest category is coding, where it ranks 38th. API pricing starts at $0.75 per million input tokens and $2.25 per million output tokens, with a 1.05M-token context window.

How much does Hy4 preview cost?

Hy4 preview costs $0.75 per million input tokens and $2.25 per million output tokens on openrouter, with cached input at $0.0378.

What is Hy4 preview's context window?

Hy4 preview accepts up to 1.05M tokens of input and can write up to 64K tokens in one response.

Is Hy4 preview open source?

Yes. Hy4 preview's weights are downloadable from Hugging Face (tencent/Hy4-preview); check the license for commercial terms.

What are Hy4 preview's strengths and weaknesses?

Relative to other ranked models, Hy4 preview places best in coding and lowest in reasoning.

What is Hy4 preview best at?

Its best category is coding, where it ranks 38th on Noometry.