PH

lora

PhiGGUF by ndavidsonUpdated 2024-05-23
Download on Hugging Face
Parameters3.82B
Context4K
Layers32
Architecturellama
Smallest quant2.2 GB
Quick answer

lora needs about 5.9 GB of memory at Q4_0 (a 2.2 GB file) with an 8K context window, so it fits on a 6 GB GPU.

lora is a 3.82B-parameter Phi model in GGUF format, quantized by ndavidson. Its trained context length is 4K tokens. The repository has 1 quant file, from 2.2 GB (Q4_0) to 2.2 GB (Q4_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_02.2 GB 4.3 GB
Fits Get file →

What GPU do you need for lora?

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: the best quant that fits is Q4_0 (2.2 GB file, about 5.9 GB in use).

Which lora quant should you choose?

Q4_0 (2.2 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters3.82B
Layers32
Trained context4,096 tokens
KV cache (fp16)384 KB per token

Source

Quantized byndavidson
FamilyPhi
Repositoryndavidson/lora
Licenseapache-2.0
Downloads (HF)26
Last updated2024-05-23

Frequently asked questions about lora

How much VRAM does lora need?

About 5.9 GB at Q4_0 with an 8K context window. That is the 2.2 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 lora on an 8 GB GPU?

Yes, with Q4_0: the 2.2 GB file needs about 5.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of lora GGUF?

It depends on the quant: Q4_0 is 2.2 GB.

What context length does lora support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 3.2 GB at 8K and 1.6 GB at 4K.

Where can I download lora GGUF?

The files are on Hugging Face in the ndavidson/lora repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does lora 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.

Other GGUFs of Phi-3-mini-4k-instruct-bnb-4bit

Similar models

Browse all →

More from ndavidson

All by ndavidson →

More Phi models

All Phi →

Other small models

All small models →