LLM
Hy4-preview
Tencent Hunyuan's 770B MoE preview text model. The card names no territorial exclusions; full licence text not reviewed.
Levelmodel
ClassLLM
Parameters770B
Context1,000,000
Released28 Aug 2026
Opennessopen weights permissive
Runs onmulti-GPU node
Statuspreview
Benchmark Results
vendor-reported
GPQA Diamondknowledge and reasoning
vendor-reported
92.3
050100
Score in %. Higher is better.
SWE-bench Procoding
vendor-reported
65.7
050100
Score in %. Higher is better.
Terminal-Bench 2.1agents and tool use
vendor-reported
85.4
050100
Score in %. Higher is better.
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
Identity
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 (256 routed experts, 8 active + 1 shared), Gated DSA attention, iHC
- Total parameters
- 770B sources
- Active parameters
- 49B
- Checkpoint parameters
- –
- Context length
- 1,000,000 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
- 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.
Links
DeveloperTencentbig tech, China
Tencentbig tech, ChinaModelHy4-previewTencent, LLM
Hy4-previewTencent, LLMR&D TopicsEfficiency Architectures (MoE, Sparse Attention, Long Context)Frontier Pretraining Scaling
Use CasesIndustrial Copilots and Maintenance AssistantsIndustrialClinical LLM (Frontier Proprietary)HealthcarePharma Documents, Regulatory and Life-Science R&DHealthcareAmbient Clinical ScribingHealthcare+2 moreUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs It
PT Pertamina Geothermal EnergyIn production, 14 Jul 2026
Banner HealthIn production, 11 Jan 2026CommureIn production, 11 Jan 2026FlatironIn production, 11 Jan 2026+39 moreDeployments of these use cases; none names this model.
PT Pertamina Geothermal EnergyIn production, 14 Jul 2026Banner HealthIn production, 11 Jan 2026CommureIn production, 11 Jan 2026FlatironIn production, 11 Jan 2026+39 moreDeployments of these use cases; none names this model.Organisations 1
Sources
- huggingface.co/tencent/Hy4-preview https://huggingface.co/tencent/Hy4-preview
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Hy4-Preview", https://aurion.semai.id/models/tencent-hy4-preview/, records of October 2026.