# Deploy Tencent models on dedicated GPUs

8 Tencent models ready to deploy on GPU capacity reserved for you, from 1.1B to 780B parameters, covering chat & reasoning, ocr & documents and translation. Private OpenAI-compatible endpoint, European GPUs, billed per GPU-hour.

| Model | Type | Params | Context | License | Page |
| --- | --- | --- | --- | --- | --- |
| Hy4-preview | Chat & reasoning | 780B | 1M | Apache 2.0 | https://sference.com/models/tencent/Hy4-preview.md |
| Hy3 | Chat & reasoning | 299B | 256K | Apache 2.0 | https://sference.com/models/tencent/Hy3.md |
| Hy-MT2-1.8B | Translation | 2.0B | 8K | Apache 2.0 | https://sference.com/models/tencent/Hy-MT2-1.8B.md |
| HunyuanOCR | OCR & documents | 1.1B | 128K | Custom license | https://sference.com/models/tencent/HunyuanOCR.md |
| Hunyuan-A13B-Instruct | Chat & reasoning | 80B | 128K | Custom license | https://sference.com/models/tencent/Hunyuan-A13B-Instruct.md |
| Hy-MT2-30B-A3B | Translation | 30B | 8K | Apache 2.0 | https://sference.com/models/tencent/Hy-MT2-30B-A3B.md |
| Hunyuan-MT-7B | Translation | 8.0B | 32K | Custom license | https://sference.com/models/tencent/Hunyuan-MT-7B.md |
| Hy-MT2-7B | Translation | 8.0B | 8K | Apache 2.0 | https://sference.com/models/tencent/Hy-MT2-7B.md |

## FAQ

### Which Tencent models can I deploy on sference?

8 today, most requested first: Hy4-preview, Hy3, Hy-MT2-1.8B, HunyuanOCR, Hunyuan-A13B-Instruct, Hy-MT2-30B-A3B, Hunyuan-MT-7B and Hy-MT2-7B. Any other Tencent checkpoint or a fine-tune of one can be requested with the form on this page.

### How many GPUs do Tencent models need?

It depends on the model: the lineup runs from 1.1B to 780B parameters. Each model page lists the checkpoint size and the minimum footprint on 80 GB and 141 GB GPUs. You don't pick hardware on sference: you choose a performance tier and we size the deployment.

### Can I use Tencent models commercially?

5 of the 8 listed models are under licenses that allow commercial use (Apache 2.0 and Custom license across the lineup). Each model page summarizes its license and links the full text.

### How is a dedicated deployment priced?

Per GPU-hour on a monthly commitment, sized to your throughput and latency targets.

Request a deployment: https://sference.com/models/tencent#request. All makers: https://sference.com/dedicated.md