LL

Meta_Llama_3.1_8b_it

LlamaGGUF by NexesenexUpdated 2024-08-20
Download on Hugging Face
Parameters8.03B
Context128K
Layers32
Architecturellama
Smallest quant2.9 GB
Quick answer

Meta_Llama_3.1_8b_it needs about 5.2 GB of memory at IQ3_S (a 3.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 10 quants, from 2.9 GB (IQ2_M) to 4.5 GB (IQ4_XS). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Meta_Llama_3.1_8b_it is a 8.03B-parameter Llama model in GGUF format, quantized by Nexesenex. Its trained context length is 128K tokens. The repository has 10 quant files, from 2.9 GB (IQ2_M) to 4.5 GB (IQ4_XS).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_M2.9 GB 4.5 GB
Fits Get file →
—3.2 GB 4.7 GB
Fits Get file →
IQ3_XXS3.3 GB 4.9 GB
Fits Get file →
IQ3_XS3.5 GB 5.1 GB
Fits Get file →
IQ3_S3.7 GB 5.2 GB
Fits Get file →
IQ3_M3.8 GB 5.3 GB
Fits Get file →
—4.0 GB 5.6 GB
Fits Get file →
—4.1 GB 5.7 GB
Fits Get file →
—4.3 GB 5.8 GB
Fits Get file →
IQ4_XS4.5 GB 6.0 GB
Fits Get file →

What GPU do you need for Meta_Llama_3.1_8b_it?

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 IQ4_XS (4.5 GB file, about 6.0 GB in use).

Which Meta_Llama_3.1_8b_it quant should you choose?

IQ3_S (3.7 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

If memory is tight, the smallest file is IQ2_M at 2.9 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters8.03B
Layers32
Trained context131,072 tokens
KV cache (fp16)128 KB per token

Source

Quantized byNexesenex
FamilyLlama
Licensellama3.1
Downloads (HF)193
Last updated2024-08-20

Frequently asked questions about Meta_Llama_3.1_8b_it

How much VRAM does Meta_Llama_3.1_8b_it need?

About 5.2 GB at IQ3_S with an 8K context window. That is the 3.7 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 Meta_Llama_3.1_8b_it on an 8 GB GPU?

Yes, with IQ4_XS: the 4.5 GB file needs about 6.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Meta_Llama_3.1_8b_it GGUF?

It depends on the quant: IQ2_M is 2.9 GB, file is 3.2 GB, IQ3_XXS is 3.3 GB, IQ3_XS is 3.5 GB, IQ3_S is 3.7 GB, IQ3_M is 3.8 GB and 4 more.

What context length does Meta_Llama_3.1_8b_it 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 Meta_Llama_3.1_8b_it GGUF?

The files are on Hugging Face in the Nexesenex/Meta_Llama-3.1-8b-it_iMat_Custom_Quant_Stategies-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 Meta_Llama_3.1_8b_it use?

Hugging Face lists the license as llama3.1. 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 Nexesenex

All by Nexesenex →

More Llama models

All Llama →

Other small models

All small models →