PY

Pythia-Chat-Base-7B

PythiaGGUF by ZeroWwUpdated 2024-06-25
Download on Hugging Face
Parameters6.86B
Context2K
Layers32
Architecturegptneox
Smallest quant5.6 GB
Quick answer

Pythia-Chat-Base-7B needs about 10.9 GB of memory at Q6_K (a 6.1 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 4 quants, from 5.6 GB (Q5_K) to 13.7 GB (F16).

Pythia-Chat-Base-7B is a 6.86B-parameter Pythia model in GGUF format, quantized by ZeroWw. Its trained context length is 2K tokens. The repository has 4 quant files, from 5.6 GB (Q5_K) to 13.7 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q5_K5.6 GB 7.2 GB
Fits Get file →
Q6_K6.1 GB 7.7 GB
Tight Get file →
Q8_07.7 GB 9.3 GB
Too big Get file →
F1613.7 GB 15.3 GB
Too big Get file →

What GPU do you need for Pythia-Chat-Base-7B?

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: the best quant that fits is Q6_K (6.1 GB file, about 10.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (7.7 GB file, about 12.5 GB in use).
  • 24 GB VRAM: the best quant that fits is F16 (13.7 GB file, about 18.5 GB in use).

Which Pythia-Chat-Base-7B quant should you choose?

Q6_K (6.1 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 Q5_K at 5.6 GB, with the quality trade-offs that come with fewer bits. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegptneox
Parameters6.86B
Layers32
Trained context2,048 tokens
KV cache (fp16)512 KB per token

Source

Quantized byZeroWw
FamilyPythia
Licensemit
Downloads (HF)31
Last updated2024-06-25

Frequently asked questions about Pythia-Chat-Base-7B

How much VRAM does Pythia-Chat-Base-7B need?

About 10.9 GB at Q6_K with an 8K context window. That is the 6.1 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 Pythia-Chat-Base-7B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 10.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Pythia-Chat-Base-7B GGUF?

It depends on the quant: Q5_K is 5.6 GB, Q6_K is 6.1 GB, Q8_0 is 7.7 GB, F16 is 13.7 GB.

What context length does Pythia-Chat-Base-7B support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 4.3 GB at 8K and 1.1 GB at 2K.

Where can I download Pythia-Chat-Base-7B GGUF?

The files are on Hugging Face in the ZeroWw/Pythia-Chat-Base-7B-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 Pythia-Chat-Base-7B 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 Pythia models

All Pythia →

Other small models

All small models →