SU

Sushi v1.3

SushiGGUF by Undi95Updated 2024-12-22
Download on Hugging Face
Parameters70.6B
Context128K
Layers80
Architecturellama
Smallest quant42.5 GB
Quick answer

Sushi v1.3 needs about 45.7 GB of memory at Q4_K_M (a 42.5 GB file) with an 8K context window, so it fits on a 48 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.

Sushi v1.3 is a 70.6B-parameter Sushi model in GGUF format, quantized by Undi95. Its trained context length is 128K tokens. The repository has 1 quant file, from 42.5 GB (Q4_K_M) to 42.5 GB (Q4_K_M).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M42.5 GB 45.7 GB
Too big Get file →

What GPU do you need for Sushi v1.3?

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: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: the best quant that fits is Q4_K_M (42.5 GB file, about 45.7 GB in use).

Which Sushi v1.3 quant should you choose?

Q4_K_M is the usual starting point for Sushi v1.3: it balances file size and quality well. Here it is a 42.5 GB file.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters70.6B
Layers80
Trained context131,072 tokens
KV cache (fp16)320 KB per token

Source

Quantized byUndi95
FamilySushi
Downloads (HF)15
Last updated2024-12-22

Frequently asked questions about Sushi v1.3

How much VRAM does Sushi v1.3 need?

About 45.7 GB at Q4_K_M with an 8K context window. That is the 42.5 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 Sushi v1.3 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 45.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Sushi v1.3 GGUF?

It depends on the quant: Q4_K_M is 42.5 GB.

What context length does Sushi v1.3 support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.

Where can I download Sushi v1.3 GGUF?

The files are on Hugging Face in the Undi95/Sushi-v1.3-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 Undi95

All by Undi95 →

More Sushi models

All Sushi →

Other large models

All large models →