LL

Llama 3.1 Minitron 4B Width Base

LlamaGGUF by ZeroWwUpdated 2024-08-17
Download on Hugging Face
Parameters4.51B
Context128K
Layers32
Architecturellama
Smallest quant3.1 GB
Quick answer

Llama 3.1 Minitron 4B Width Base needs about 6.2 GB of memory at Q6_K (a 4.6 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 6 quants, from 3.1 GB to 9.0 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Llama 3.1 Minitron 4B Width Base is a 4.51B-parameter Llama model in GGUF format, quantized by ZeroWw. Its trained context length is 128K tokens. The repository has 6 quant files, from 3.1 GB to 9.0 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—3.1 GB 4.6 GB
Fits Get file →
Q5_K4.2 GB 5.8 GB
Fits Get file →
Q6_K4.6 GB 6.2 GB
Fits Get file →
—4.8 GB 6.4 GB
Fits Get file →
Q8_05.5 GB 7.1 GB
Fits Get file →
F169.0 GB 10.6 GB
Too big Get file →

What GPU do you need for Llama 3.1 Minitron 4B Width Base?

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 Q5_K (4.2 GB file, about 5.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (5.5 GB file, about 7.1 GB in use).
  • 12 GB VRAM: the best quant that fits is F16 (9.0 GB file, about 10.6 GB in use).

Which Llama 3.1 Minitron 4B Width Base quant should you choose?

Q6_K (4.6 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.

If memory is tight, the smallest file is the first one in the table at 3.1 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters4.51B
Layers32
Trained context131,072 tokens
KV cache (fp16)128 KB per token

Source

Quantized byZeroWw
FamilyLlama
Licensemit
Downloads (HF)45
Last updated2024-08-17

Frequently asked questions about Llama 3.1 Minitron 4B Width Base

How much VRAM does Llama 3.1 Minitron 4B Width Base need?

About 6.2 GB at Q6_K with an 8K context window. That is the 4.6 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 Llama 3.1 Minitron 4B Width Base on an 8 GB GPU?

Yes, with Q8_0: the 5.5 GB file needs about 7.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Llama 3.1 Minitron 4B Width Base GGUF?

It depends on the quant: file is 3.1 GB, Q5_K is 4.2 GB, Q6_K is 4.6 GB, file is 4.8 GB, Q8_0 is 5.5 GB, F16 is 9.0 GB.

What context length does Llama 3.1 Minitron 4B Width Base support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download Llama 3.1 Minitron 4B Width Base GGUF?

The files are on Hugging Face in the ZeroWw/Llama-3.1-Minitron-4B-Width-Base-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 Llama 3.1 Minitron 4B Width Base use?

Hugging Face lists the license as mit. 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 ZeroWw

All by ZeroWw →

More Llama models

All Llama →

Other small models

All small models →