LL

codellama

LlamaGGUF by xaskasdfUpdated 2024-07-03
Download on Hugging Face
Parameters13B
Context16K
Layers40
Architecturellama
Smallest quant13.8 GB
Quick answer

codellama needs about 21.0 GB of memory at Q8_0 (a 13.8 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 2 quants, from 13.8 GB (Q8_0) to 26.0 GB (F16).

codellama is a 13B-parameter Llama model in GGUF format, quantized by xaskasdf. Its trained context length is 16K tokens. The repository has 2 quant files, from 13.8 GB (Q8_0) to 26.0 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_013.8 GB 21.0 GB
Too big Get file →
F1626.0 GB 33.2 GB
Too big Get file →

What GPU do you need for codellama?

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 Q8_0 (13.8 GB file, about 21.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (13.8 GB file, about 21.0 GB in use).
  • 48 GB VRAM: the best quant that fits is F16 (26.0 GB file, about 33.2 GB in use).

Which codellama quant should you choose?

Q8_0 (13.8 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters13B
Layers40
Trained context16,384 tokens
KV cache (fp16)800 KB per token

Source

Quantized byxaskasdf
FamilyLlama
Downloads (HF)23
Last updated2024-07-03

Frequently asked questions about codellama

How much VRAM does codellama need?

About 21.0 GB at Q8_0 with an 8K context window. That is the 13.8 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 codellama on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 21.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of codellama GGUF?

It depends on the quant: Q8_0 is 13.8 GB, F16 is 26.0 GB.

What context length does codellama support?

The GGUF header lists a trained context length of 16,384 tokens (16K). The KV cache grows with the context you use: about 6.7 GB at 8K and 13.4 GB at 16K.

Where can I download codellama GGUF?

The files are on Hugging Face in the xaskasdf/codellama-python-13b-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Similar models

Browse all →

More from xaskasdf

All by xaskasdf →

More Llama models

All Llama →

Other medium models

All medium models →