Context
262K tokens (~131 books)
Input $/1M
$0.06
Output $/1M
$0.21
Type
text
License
Open Source
Benchmarks
1 tested
Data as of
About
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...
Tested on 1 benchmarks. Top scores: Artificial Analysis — Quality Index (33.6%).
Capabilities
speed
33.6
#78 globally
Benchmark Scores
Compare AllTested on 1 benchmarks · Ranked across 1 categories
Score Distribution (all 312 models)
0255075100
speedCompare speed →
Artificial Analysis — Quality Index
33.6·Artificial Analysis Quality Index. Composite quality score combining multiple benchmark results into a single metric.
Quick compare:
vs Qwen3 Coder Next FP8
vs Qwen2.5 Coder 32B Instruct AWQ
P
vs Speaker Diarization Community 1 Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Research
Documentation
Community
Source Code
BenchGecko API
hy3-preview
Specifications
- Typetext
- Context262K tokens (~131 books)
- ReleasedApr 2026
- LicenseOpen Source
- Statuspreview
- Cost / Message~$0.000
Available On
Categories
Share & Export
Frequently Asked Questions
Hy3 preview is an open-source text AI model by tencent, released in April 2026. Context window: 262K tokens.
Related Models
Qwen3 Coder Next FP8 · AlibabaQwen2.5 Coder 32B Instruct AWQ · AlibabaSpeaker Diarization Community 1 · PyannoteGemma 3 12B (free) · Google DeepMindQwen3 Next 80B A3B Instruct (free) · Alibaba QwenBenchmarks
Artificial Analysis — Quality IndexKey facts · as of 2026-06-22
- Hy3 preview by tencent. Not enough public benchmark scores to rank yet.
- List price $0.0630 input · $0.21 output per 1M tokens (as of 2026-06-22).
How to cite · data as of 2026-06-22
Hy3 preview · benchmarks, pricing and providers. BenchGecko, data as of 2026-06-22. https://benchgecko.ai/model/hy3-preview
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP