UL

ultra.star

UltraGGUF by liminerityUpdated 2024-03-29
Download on Hugging Face
Parameters1.83B
Context16K
Layers24
Architecturellama
Smallest quant1.9 GB
Quick answer

ultra.star needs about 3.0 GB of memory at Q8_0 (a 1.9 GB file) with an 8K context window, so it fits on a 4 GB GPU.

ultra.star is a 1.83B-parameter Ultra model in GGUF format, quantized by liminerity. Its trained context length is 16K tokens. The repository has 1 quant file, from 1.9 GB (Q8_0) to 1.9 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_01.9 GB 3.0 GB
Fits Get file →

What GPU do you need for ultra.star?

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 Q8_0 (1.9 GB file, about 3.0 GB in use).

Which ultra.star quant should you choose?

Q8_0 (1.9 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters1.83B
Layers24
Trained context16,384 tokens
KV cache (fp16)60 KB per token

Source

Quantized byliminerity
FamilyUltra
Licenseapache-2.0
Downloads (HF)11
Last updated2024-03-29

Frequently asked questions about ultra.star

How much VRAM does ultra.star need?

About 3.0 GB at Q8_0 with an 8K context window. That is the 1.9 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 ultra.star on an 8 GB GPU?

Yes, with Q8_0: the 1.9 GB file needs about 3.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of ultra.star GGUF?

It depends on the quant: Q8_0 is 1.9 GB.

What context length does ultra.star support?

The GGUF header lists a trained context length of 16,384 tokens (16K). The KV cache grows with the context you use: about 0.5 GB at 8K and 1.0 GB at 16K.

Where can I download ultra.star GGUF?

The files are on Hugging Face in the liminerity/ultra.star.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 ultra.star 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.

Similar models

Browse all →

More from liminerity

All by liminerity →

More Ultra models

All Ultra →

Other tiny models

All tiny models →