GenericMoEUltra 3B A1B needs about 2.0 GB of memory at its main file (a 1.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.
GenericMoEUltra 3B A1B is a 1.07B-parameter Genericmoeultra model in GGUF format, quantized by Hsh1662. Its trained context length is 164K tokens. The repository has 1 quant file, from 1.1 GB to 1.1 GB.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 1.1 GB | 2.0 GB | Fits | Get file → |
What GPU do you need for GenericMoEUltra 3B A1B?
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 (1.1 GB file, about 2.0 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about GenericMoEUltra 3B A1B
How much VRAM does GenericMoEUltra 3B A1B need?
About 2.0 GB at its main file with an 8K context window. That is the 1.1 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 GenericMoEUltra 3B A1B on an 8 GB GPU?
Yes, with the smaller files: the 1.1 GB file needs about 2.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of GenericMoEUltra 3B A1B GGUF?
It depends on the quant: file is 1.1 GB.
What context length does GenericMoEUltra 3B A1B support?
The GGUF header lists a trained context length of 168,000 tokens (164K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.
Where can I download GenericMoEUltra 3B A1B GGUF?
The files are on Hugging Face in the Hsh1662/GenericMoEUltra-3B-A1B repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.