GE

saiga_gemma_9b

GemmaGGUF by NightFoxUpdated 2024-03-01
Download on Hugging Face
Parameters8.54B
Context8K
Layers28
Architecturegemma
Smallest quant6.1 GB
Quick answer

saiga_gemma_9b needs about 11.3 GB of memory at Q6_K (a 7.0 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 3 quants, from 6.1 GB (Q5_K_M) to 9.1 GB (Q8_0).

saiga_gemma_9b is a 8.54B-parameter Gemma model in GGUF format, quantized by NightFox. Its trained context length is 8K tokens. The repository has 3 quant files, from 6.1 GB (Q5_K_M) to 9.1 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q5_K_M6.1 GB 10.4 GB
Too big Get file →
Q6_K7.0 GB 11.3 GB
Too big Get file →
Q8_09.1 GB 13.3 GB
Too big Get file →

What GPU do you need for saiga_gemma_9b?

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 (7.0 GB file, about 11.3 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (9.1 GB file, about 13.3 GB in use).

Which saiga_gemma_9b quant should you choose?

Q6_K (7.0 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_M at 6.1 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

Architecturegemma
Parameters8.54B
Layers28
Trained context8,192 tokens
KV cache (fp16)448 KB per token

Source

Quantized byNightFox
FamilyGemma
Downloads (HF)89
Last updated2024-03-01

Frequently asked questions about saiga_gemma_9b

How much VRAM does saiga_gemma_9b need?

About 11.3 GB at Q6_K with an 8K context window. That is the 7.0 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 saiga_gemma_9b 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 saiga_gemma_9b GGUF?

It depends on the quant: Q5_K_M is 6.1 GB, Q6_K is 7.0 GB, Q8_0 is 9.1 GB.

What context length does saiga_gemma_9b support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 3.8 GB at 8K and 3.8 GB at 8K.

Where can I download saiga_gemma_9b GGUF?

The files are on Hugging Face in the NightFox/saiga_gemma_9b_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 NightFox

All by NightFox →

More Gemma models

All Gemma →

Other small models

All small models →