tencent.Hy3 needs about 181.8 GB of memory at F16 (a 178.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 6 quants, from 107.4 GB (F16) to 313.7 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
tencent.Hy3 is a 295B-parameter model in GGUF format, quantized by DevQuasar. Its trained context length is 256K tokens. The repository has 6 quant files, from 107.4 GB (F16) to 313.7 GB (F16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| F16 | 107.4 GB (8 parts) | 110.6 GB | Too big | Get file → | |
| F16 | 141.0 GB (11 parts) | 144.1 GB | Too big | Get file → | |
| F16 | 178.6 GB (13 parts) | 181.8 GB | Too big | Get file → | |
| F16 | 209.4 GB (16 parts) | 212.6 GB | Too big | Get file → | |
| F16 | 242.2 GB (19 parts) | 245.4 GB | Too big | Get file → | |
| F16 | 313.7 GB (24 parts) | 316.8 GB | Too big | Get file → |
What GPU do you need for tencent.Hy3?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which tencent.Hy3 quant should you choose?
F16 (178.6 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
If memory is tight, the smallest file is F16 at 107.4 GB, with the quality trade-offs that come with fewer bits. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about tencent.Hy3
How much VRAM does tencent.Hy3 need?
About 181.8 GB at F16 with an 8K context window. That is the 178.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run tencent.Hy3 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 110.6 GB, so you would need a larger GPU or CPU offloading.
What is the file size of tencent.Hy3 GGUF?
It depends on the quant: F16 is 107.4 GB, F16 is 141.0 GB, F16 is 178.6 GB, F16 is 209.4 GB, F16 is 242.2 GB, F16 is 313.7 GB.
What context length does tencent.Hy3 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.
Where can I download tencent.Hy3 GGUF?
The files are on Hugging Face in the DevQuasar/tencent.Hy3-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does tencent.Hy3 use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.