MO

MobileLLM-600M

MobilellmGGUF by pjh64Updated 2024-10-31
Download on Hugging Face
Parameters640M
Context2K
Layers40
Architecturellama
Smallest quant0.4 GB
Quick answer

MobileLLM-600M needs about 1.5 GB of memory at Q4_K_M (a 0.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 13 quants, from 0.4 GB (Q4_0) to 2.6 GB (F32).

MobileLLM-600M is a 640M-parameter Mobilellm model in GGUF format, quantized by pjh64. Its trained context length is 2K tokens. The repository has 13 quant files, from 0.4 GB (Q4_0) to 2.6 GB (F32).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_00.4 GB 1.0 GB
Fits Get file →
Q4_10.4 GB 1.0 GB
Fits Get file →
Q4_K_S0.4 GB 1.1 GB
Fits Get file →
Q5_00.5 GB 1.1 GB
Fits Get file →
Q4_K0.5 GB 1.1 GB
Fits Get file →
Q4_K_M0.5 GB 1.1 GB
Fits Get file →
Q5_K_S0.5 GB 1.1 GB
Fits Get file →
Q5_10.5 GB 1.1 GB
Fits Get file →
Q5_K0.5 GB 1.1 GB
Fits Get file →
Q5_K_M0.5 GB 1.1 GB
Fits Get file →
Q6_K0.6 GB 1.3 GB
Fits Get file →
Q8_00.7 GB 1.3 GB
Fits Get file →
F322.6 GB 3.2 GB
Fits Get file →

What GPU do you need for MobileLLM-600M?

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 F32 (2.6 GB file, about 3.6 GB in use).

Which MobileLLM-600M quant should you choose?

Q4_K_M is the usual starting point for MobileLLM-600M: it balances file size and quality well. Here it is a 0.5 GB file.

The Q8_0 file is 0.7 GB, about 1.5 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q4_0 at 0.4 GB, with the quality trade-offs that come with fewer bits. 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
Parameters640M
Layers40
Trained context2,048 tokens
KV cache (fp16)60 KB per token

Source

Quantized bypjh64
FamilyMobilellm
Downloads (HF)57
Last updated2024-10-31

Frequently asked questions about MobileLLM-600M

How much VRAM does MobileLLM-600M need?

About 1.5 GB at Q4_K_M with an 8K context window. That is the 0.5 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 MobileLLM-600M on an 8 GB GPU?

Yes, with F32: the 2.6 GB file needs about 3.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of MobileLLM-600M GGUF?

It depends on the quant: Q4_0 is 0.4 GB, Q4_1 is 0.4 GB, Q4_K_S is 0.4 GB, Q5_0 is 0.5 GB, Q4_K is 0.5 GB, Q4_K_M is 0.5 GB and 7 more.

What context length does MobileLLM-600M support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 0.5 GB at 8K and 0.1 GB at 2K.

Where can I download MobileLLM-600M GGUF?

The files are on Hugging Face in the pjh64/MobileLLM-600M-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 MobileLLM-600M 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.

Similar models

Browse all →

More from pjh64

All by pjh64 →

More Mobilellm models

All Mobilellm →

Other tiny models

All tiny models →