Project AURIONAI Use-Cases Research & Innovation
Tencent
LLM

Hy3

Tencent Hunyuan's text-only open-weight MoE model.

TencentTencentChina
Figure from the Hy3 model card.
Levelmodel
ClassLLM
Parameters295B
Context256,000
Released6 Jul 2026
Opennessopen weights permissive
Runs onmulti-GPU node
Statusactive

From the model's own page; who made each picture and where it comes from are listed on the Image Credits page.

Benchmark Results

vendor-reportedindependent
Arena Text leaderboardhuman preference arena
independent
no score in the notes
rank 73, Arena Text overall8 Oct 2026 arena.ai
Artificial Analysis Intelligence Index v4.3.2composite index
independent
25
03060

Score in index. Higher is better.

AA leaderboard, 205 current models10 Oct 2026 artificialanalysis.ai
GPQA Diamondknowledge and reasoning
vendor-reported
90.4
050100

Score in %. Higher is better.

huggingface.co
SWE-bench Verifiedcoding
vendor-reported
78
050100

Score in %. Higher is better.

huggingface.co
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).

R&D Topics

Efficiency architectures (MoE, sparse attention, long context)produced byOpen-weight frontier releases as research substrateused by

Identity

Developer
Tencent
Family
not in the notes
Released
6 Jul 2026 sourcessecondary report
Weights published
–
Record checked
10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
MoE MoE (192 experts, top-8 active) plus a 3.8B MTP layer
Total parameters
295B sources
Active parameters
21B
Checkpoint parameters
298.79B
Note
Plus a 3.8B MTP layer, which brings the checkpoint to about 299B
Context length
256,000 tokens
Input
text
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
Apache License 2.0 sources
Access routes
API, open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
multi-GPU node
Footprint
FP8 variant
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Lineage

The notes record no lineage for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. huggingface.co/tencent/Hy3/raw/main/README.md https://huggingface.co/tencent/Hy3/raw/main/README.md
  2. huggingface.co/tencent/Hy3/raw/main/LICENSE https://huggingface.co/tencent/Hy3/raw/main/LICENSE
  3. huggingface.co/api/models/tencent/Hy3/commits/main https://huggingface.co/api/models/tencent/Hy3/commits/main
  4. 腾讯混元Hy3发布：Agent能力和产品体验跃升-新华网 news.cn https://www.news.cn/tech/20260706/d611426ecba54475b2dd94920e4e3557/c.html
  5. Hy3 - Intelligence, Performance & Price Analysis | Artificial Analysis artificialanalysis.ai https://artificialanalysis.ai/models/hy3
  6. Text Arena: LLM Rankings & AI Model Benchmarks arena.ai https://arena.ai/leaderboard/chat/text
  7. huggingface.co/tencent/Hy3 https://huggingface.co/tencent/Hy3

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Hy3", https://aurion.semai.id/models/tencent-hy3/, records of October 2026.