SM

Smaug 72B v0.1

SmaugGGUF by senseableUpdated 2024-02-05
Download on Hugging Face
Parameters72.3B
Context32K
Layers80
Architecturellama
Smallest quant30.4 GB
Quick answer

Smaug 72B v0.1 needs about 65.7 GB of memory at Q4_K_M (a 43.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 3 quants, from 30.4 GB (Q2_K) to 49.9 GB (Q5_K_S). Longer context needs more: at 32K tokens the KV cache alone adds about 85.9 GB.

Smaug 72B v0.1 is a 72.3B-parameter Smaug model in GGUF format, quantized by senseable. Its trained context length is 32K tokens. The repository has 3 quant files, from 30.4 GB (Q2_K) to 49.9 GB (Q5_K_S).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K30.4 GB 52.3 GB
Too big Get file →
Q4_K_M43.8 GB 65.7 GB
Too big Get file →
Q5_K_S49.9 GB 71.9 GB
Too big Get file →

What GPU do you need for Smaug 72B v0.1?

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: no quant fits at an 8K context.

Which Smaug 72B v0.1 quant should you choose?

Q4_K_M is the usual starting point for Smaug 72B v0.1: it balances file size and quality well. Here it is a 43.8 GB file.

If memory is tight, the smallest file is Q2_K at 30.4 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters72.3B
Layers80
Trained context32,768 tokens
KV cache (fp16)2560 KB per token

Source

Quantized bysenseable
FamilySmaug
Downloads (HF)68
Last updated2024-02-05

Frequently asked questions about Smaug 72B v0.1

How much VRAM does Smaug 72B v0.1 need?

About 65.7 GB at Q4_K_M with an 8K context window. That is the 43.8 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 Smaug 72B v0.1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 52.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Smaug 72B v0.1 GGUF?

It depends on the quant: Q2_K is 30.4 GB, Q4_K_M is 43.8 GB, Q5_K_S is 49.9 GB.

What context length does Smaug 72B v0.1 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 21.5 GB at 8K and 85.9 GB at 32K.

Where can I download Smaug 72B v0.1 GGUF?

The files are on Hugging Face in the senseable/Smaug-72B-v0.1-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 senseable

All by senseable →

More Smaug models

All Smaug →

Other large models

All large models →