Hermyale Stack 90B needs about 44.1 GB of memory at IQ3_S (a 40.0 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 8 quants, from 34.1 GB (Q2_K) to 49.8 GB (IQ4_XS). Longer context needs more: at 32K tokens the KV cache alone adds about 14.1 GB.
Hermyale Stack 90B is a 91.9B-parameter Hermyale model in GGUF format, quantized by mradermacher. Its trained context length is 128K tokens. The repository has 8 quant files, from 34.1 GB (Q2_K) to 49.8 GB (IQ4_XS).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 34.1 GB | 38.2 GB | Too big | Get file → | |
| IQ3_XS | 38.0 GB | 42.0 GB | Too big | Get file → | |
| Q3_K_S | 39.9 GB | 44.0 GB | Too big | Get file → | |
| IQ3_S | 40.0 GB | 44.1 GB | Too big | Get file → | |
| IQ3_M | 41.4 GB | 45.4 GB | Too big | Get file → | |
| Q3_K_M | 44.4 GB | 48.5 GB | Too big | Get file → | |
| Q3_K_L | 48.3 GB | 52.4 GB | Too big | Get file → | |
| IQ4_XS | 49.8 GB | 53.8 GB | Too big | Get file → |
What GPU do you need for Hermyale Stack 90B?
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: the best quant that fits is IQ3_M (41.4 GB file, about 45.4 GB in use).
Which Hermyale Stack 90B quant should you choose?
IQ3_S (40.0 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
If memory is tight, the smallest file is Q2_K at 34.1 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Hermyale Stack 90B
How much VRAM does Hermyale Stack 90B need?
About 44.1 GB at IQ3_S with an 8K context window. That is the 40.0 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 Hermyale Stack 90B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 38.2 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Hermyale Stack 90B GGUF?
It depends on the quant: Q2_K is 34.1 GB, IQ3_XS is 38.0 GB, Q3_K_S is 39.9 GB, IQ3_S is 40.0 GB, IQ3_M is 41.4 GB, Q3_K_M is 44.4 GB and 2 more.
What context length does Hermyale Stack 90B support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 3.5 GB at 8K and 14.1 GB at 32K.
Where can I download Hermyale Stack 90B GGUF?
The files are on Hugging Face in the mradermacher/Hermyale-stack-90B-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 Hermyale Stack 90B use?
Hugging Face lists the license as cc-by-nc-4.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.