Hy3 needs about 185.4 GB of memory at Q4_K_M (a 182.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 0.6 GB to 317.7 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 10.9 GB.
Hy3 – 300B‑parameter Mixture‑of‑Experts model
Hy3 is a 300‑billion‑parameter Mixture‑of‑Experts model with 21 billion active parameters. It was trained on 262 144 context tokens and released under the Apache‑2.0 license. The model is available in multiple GGUF quantizations.
Good at
- Large‑context reasoning
- Agentic task performance
- Long‑form content generation
Limits
- Requires high VRAM for larger quants
- No vision support
- No official benchmark scores provided
Best for
- Coding assistance
- Productivity workflows
- Agent‑based applications
Who it’s for. Developers and researchers needing a large‑scale, Mixture‑of‑Experts model for text‑only tasks. The model is suitable for users who can manage the memory requirements of the larger quantizations.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.6 GB | 3.8 GB | Fits | Get file → | |
| IQ1_S | 64.1 GB (2 parts) | 67.3 GB | Too big | Get file → | |
| IQ1_M | 71.1 GB (2 parts) | 74.3 GB | Too big | Get file → | |
| IQ2_XXS | 82.1 GB (3 parts) | 85.3 GB | Too big | Get file → | |
| IQ2_XS | 91.1 GB (3 parts) | 94.3 GB | Too big | Get file → | |
| IQ2_S | 92.8 GB (3 parts) | 96.0 GB | Too big | Get file → | |
| IQ2_M | 102.1 GB (3 parts) | 105.3 GB | Too big | Get file → | |
| Q2_K | 106.7 GB (3 parts) | 109.9 GB | Too big | Get file → | |
| Q2_K_L | 107.2 GB (3 parts) | 110.4 GB | Too big | Get file → | |
| IQ3_XXS | 126.1 GB (4 parts) | 129.3 GB | Too big | Get file → | |
| Q3_K_S | 131.3 GB (4 parts) | 134.5 GB | Too big | Get file → | |
| IQ3_XS | 137.4 GB (4 parts) | 140.7 GB | Too big | Get file → | |
| Q3_K_M | 137.5 GB (4 parts) | 140.8 GB | Too big | Get file → | |
| Q3_K_L | 143.1 GB (4 parts) | 146.3 GB | Too big | Get file → | |
| Q3_K_XL | 143.5 GB (4 parts) | 146.7 GB | Too big | Get file → | |
| IQ3_M | 143.7 GB (4 parts) | 146.9 GB | Too big | Get file → | |
| IQ4_XS | 160.8 GB (5 parts) | 164.0 GB | Too big | Get file → | |
| IQ4_NL | 169.8 GB (5 parts) | 173.0 GB | Too big | Get file → | |
| Q4_0 | 170.3 GB (5 parts) | 173.5 GB | Too big | Get file → | |
| Q4_K_S | 175.5 GB (5 parts) | 178.7 GB | Too big | Get file → | |
| Q4_K_M | 182.2 GB (5 parts) | 185.4 GB | Too big | Get file → | |
| Q4_K_L | 182.5 GB (5 parts) | 185.7 GB | Too big | Get file → | |
| Q4_1 | 187.8 GB (5 parts) | 191.0 GB | Too big | Get file → | |
| Q5_K_S | 206.0 GB (6 parts) | 209.3 GB | Too big | Get file → | |
| Q5_K_M | 212.8 GB (6 parts) | 216.1 GB | Too big | Get file → | |
| Q6_K | 257.2 GB (7 parts) | 260.5 GB | Too big | Get file → | |
| Q8_0 | 317.7 GB (9 parts) | 320.9 GB | Too big | Get file → |
For 8 GB VRAM, use the unknown quant (0.6 GB file, 3.8 GB memory). For 16 GB VRAM, IQ1_S (64.05 GB file, 67.3 GB memory) or IQ1_M (71.1 GB file, 74.3 GB memory) are options. For 32 GB VRAM, Q4_K_M (182.16 GB file, 185.4 GB memory) or Q5_K_M (212.84 GB file, 216.1 GB memory) provide higher quality. Memory estimates are approximate.
What GPU do you need for 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: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
- 48 GB VRAM: the best quant that fits is the main file (0.6 GB file, about 3.8 GB in use).
Which Hy3 quant should you choose?
Q4_K_M is the usual starting point for Hy3: it balances file size and quality well. Here it is a 182.2 GB file.
The Q8_0 file is 317.7 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is the first one in the table at 0.6 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Hy3
How much VRAM does Hy3 need?
About 185.4 GB at Q4_K_M with an 8K context window. That is the 182.2 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 Hy3 on an 8 GB GPU?
Yes, with the smaller files: the 0.6 GB file needs about 3.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Hy3 GGUF?
It depends on the quant: file is 0.6 GB, IQ1_S is 64.1 GB, IQ1_M is 71.1 GB, IQ2_XXS is 82.1 GB, IQ2_XS is 91.1 GB, IQ2_S is 92.8 GB and 21 more.
What context length does 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 10.9 GB at 32K.
Where can I download Hy3 GGUF?
The files are on Hugging Face in the bartowski/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 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.
Can I run Hy3 on 8 GB VRAM?
Yes, the unknown quant file is 0.6 GB and requires about 3.8 GB of VRAM at 8 k context.
Does Hy3 support vision tasks?
No, Hy3 is a text‑only model with no vision capabilities.