LL

llama323Btuned

LlamaGGUF by KevinkreUpdated 2024-09-27
Download on Hugging Face
Parameters3.21B
Context128K
Layers28
Architecturellama
Smallest quant2.6 GB
Quick answer

llama323Btuned needs about 4.1 GB of memory at Q6_K (a 2.6 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 2 quants, from 2.6 GB (Q6_K) to 6.4 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.

llama323Btuned is a 3.21B-parameter Llama model in GGUF format, quantized by Kevinkre. Its trained context length is 128K tokens. The repository has 2 quant files, from 2.6 GB (Q6_K) to 6.4 GB (BF16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q6_K2.6 GB 4.1 GB
Fits Get file →
BF166.4 GB 7.9 GB
Tight Get file →

What GPU do you need for llama323Btuned?

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 Q6_K (2.6 GB file, about 4.1 GB in use).
  • 8 GB VRAM: the best quant that fits is BF16 (6.4 GB file, about 7.9 GB in use).

Which llama323Btuned quant should you choose?

Q6_K (2.6 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.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters3.21B
Layers28
Trained context131,072 tokens
KV cache (fp16)112 KB per token

Source

Quantized byKevinkre
FamilyLlama
Licenseapache-2.0
Downloads (HF)39
Last updated2024-09-27

Frequently asked questions about llama323Btuned

How much VRAM does llama323Btuned need?

About 4.1 GB at Q6_K with an 8K context window. That is the 2.6 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 llama323Btuned on an 8 GB GPU?

Yes, with BF16: the 6.4 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of llama323Btuned GGUF?

It depends on the quant: Q6_K is 2.6 GB, BF16 is 6.4 GB.

What context length does llama323Btuned support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.

Where can I download llama323Btuned GGUF?

The files are on Hugging Face in the Kevinkre/llama323Btuned repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does llama323Btuned 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 Kevinkre

All by Kevinkre →

More Llama models

All Llama →

Other tiny models

All tiny models →