EBEasy BenchmarksLLM model index
Workspace
Overview
Benchmarks
Benchmarks list
Overall Index
Coding
Math
MMLU-Pro
Speed
Value
Models
All models
GPT-5.5 (xhigh)
GPT-5.5 (high)
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)
Gemini 3.1 Pro Preview
GPT-5.4 (xhigh)
Artificial Analysis data
Back
TE

Tencent

Hy3-preview (Reasoning)

Hy3-preview (Reasoning) is a Tencent language-model profile in the Easy Benchmarks snapshot. Use this page to compare its measured Artificial Analysis scores, output speed, time to first token, pricing, and relative ranking against other models in the local catalog.

Artificial Analysis model profile

Operational Metrics

Output Speed86.4 tok/s
First Token2.37s
Blended Price-

Model Metadata

Queryable facts extracted from the upstream model payload.

ReleaseApr 23, 2026
Context Windown/a
Modalitiesn/a
API fields: release_date

Strength: GPQA

Rank #37 across 478 models.

86.7%

Strength: HLE

Rank #45 across 474 models.

25.5%

Strength: Overall

Rank #61 across 500 models.

41.9

Watch Area: TTFT

Rank #230 across 293 models.

2.37s

Strength Profile

Percentile score by analysis domain.

Benchmark Percentiles

Higher bars mean stronger relative placement.

All Benchmarks

MetricDomainValueRank
Artificial Analysis Intelligence Indexoverall41.9#61
Artificial Analysis Coding Indexcoding36.5#69
GPQAreasoning86.7%#37
Humanity's Last Examreasoning25.5%#45
SciCode
coding, reasoning
41.2%
#72
Output Speedspeed86.4 tok/s#147
Time to First Tokenspeed2.37s#230