LL

llama fine tune

LlamaGGUF by Jai-VUpdated 2024-12-10
Download on Hugging Face
Parameters13B
Context4K
Layers40
Architecturellama
Smallest quant10.5 GB
Quick answer

llama fine tune needs about 17.8 GB of memory at its main file (a 10.5 GB file) with an 8K context window, so it fits on a 24 GB GPU.

llama fine tune is a 13B-parameter Llama model in GGUF format, quantized by Jai-V. Its trained context length is 4K tokens. The repository has 1 quant file, from 10.5 GB to 10.5 GB.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—10.5 GB 14.4 GB
Too big Get file →

What GPU do you need for llama fine tune?

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: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: the best quant that fits is the main file (10.5 GB file, about 17.8 GB in use).

Specifications

Architecture from the GGUF header

Architecturellama
Parameters13B
Layers40
Trained context4,096 tokens
KV cache (fp16)800 KB per token

Source

Quantized byJai-V
FamilyLlama
Licensellama2
Downloads (HF)6
Last updated2024-12-10

Frequently asked questions about llama fine tune

How much VRAM does llama fine tune need?

About 17.8 GB at its main file with an 8K context window. That is the 10.5 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 llama fine tune on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 17.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of llama fine tune GGUF?

It depends on the quant: file is 10.5 GB.

What context length does llama fine tune support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 6.7 GB at 8K and 3.4 GB at 4K.

Where can I download llama fine tune GGUF?

The files are on Hugging Face in the Jai-V/llama_fine_tune-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 fine tune use?

Hugging Face lists the license as llama2. 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 Jai-V

All by Jai-V →

More Llama models

All Llama →

Other medium models

All medium models →