Hy3 1M needs about 185.8 GB of memory at Q4_K_M (a 182.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 10 quants, from 7.5 GB (F16) to 245.3 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 10.9 GB.
Hy3 1M: 295B‑parameter MoE model with 1M token context
Hy3 1M is a 295B‑parameter mixture‑of‑experts model based on hy_v3 architecture. It supports a 1,048,576 token context window and is licensed under apache‑2.0. The model is optimized for reasoning and tool use.
Good at
- Large context reasoning
- Tool‑enabled tasks
- High‑parameter MoE performance
Limits
- Requires large VRAM for full precision
- No vision support
- Limited to text only
Best for
- Long‑form content generation
- Complex coding assistance
- Enterprise productivity workflows
Who it’s for. Developers and researchers needing a high‑capacity language model with extended context. The model is suitable for applications that can accommodate its memory footprint and benefit from MoE efficiency.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| F16 | 7.5 GB | 10.7 GB | Too big | Get file → | |
| IQ2_M | 96.0 GB | 99.2 GB | Too big | Get file → | |
| IQ2_M | 100.0 GB | 103.2 GB | Too big | Get file → | |
| Q2_K | 107.4 GB | 110.6 GB | Too big | Get file → | |
| Q2_K | 111.4 GB | 114.6 GB | Too big | Get file → | |
| IQ3_XXS | 117.3 GB | 120.6 GB | Too big | Get file → | |
| Q3_K_M | 144.9 GB | 148.2 GB | Too big | Get file → | |
| Q4_K_M | 182.6 GB | 185.8 GB | Too big | Get file → | |
| Q5_K_M | 213.4 GB | 216.6 GB | Too big | Get file → | |
| Q6_K | 245.3 GB (6 parts) | 248.5 GB | Too big | Get file → |
For 8 GB VRAM, use IQ2_M (file size 96.02 GB, estimated 99.2 GB memory at 8 k context). For 16 GB VRAM, Q2_K (111.38 GB, 114.6 GB memory) is a good balance. For 32 GB VRAM, Q4_K_M (182.56 GB, 185.8 GB memory) offers higher precision. Estimates are approximate.
What GPU do you need for Hy3 1M?
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: the best quant that fits is F16 (7.5 GB file, about 10.7 GB in use).
- 16 GB VRAM: the best quant that fits is F16 (7.5 GB file, about 10.7 GB in use).
- 24 GB VRAM: the best quant that fits is F16 (7.5 GB file, about 10.7 GB in use).
- 32 GB VRAM: the best quant that fits is F16 (7.5 GB file, about 10.7 GB in use).
- 48 GB VRAM: the best quant that fits is F16 (7.5 GB file, about 10.7 GB in use).
Which Hy3 1M quant should you choose?
Q4_K_M is the usual starting point for Hy3 1M: it balances file size and quality well. Here it is a 182.6 GB file.
If memory is tight, the smallest file is F16 at 7.5 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 Hy3 1M
How much VRAM does Hy3 1M need?
About 185.8 GB at Q4_K_M with an 8K context window. That is the 182.6 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 Hy3 1M on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Hy3 1M GGUF?
It depends on the quant: F16 is 7.5 GB, IQ2_M is 96.0 GB, IQ2_M is 100.0 GB, Q2_K is 107.4 GB, Q2_K is 111.4 GB, IQ3_XXS is 117.3 GB and 4 more.
What context length does Hy3 1M support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K). 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 1M GGUF?
The files are on Hugging Face in the satgeze/Hy3-1M-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 1M 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 1M on 8 GB VRAM?
No. The smallest quant requires about 99.2 GB memory at 8 k context, exceeding 8 GB VRAM.
Does Hy3 1M support vision tasks?
No. The model is text‑only and has no vision capability.
What license applies to Hy3 1M?
The model is released under the apache‑2.0 license.