MI

miquella 120b

MiquellaGGUF by alpindaleUpdated 2024-02-01
Download on Hugging Face
Parameters118B
Context32K
Layers137
Architecturellama
Smallest quant31.1 GB
Quick answer

miquella 120b needs about 45.1 GB of memory at Q2_K_S (a 40.0 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 6 quants, from 31.1 GB (IQ2_XXS) to 47.9 GB (Q3_K_XS).

miquella 120b is a 118B-parameter Miquella model in GGUF format, quantized by alpindale. Its trained context length is 32K tokens. The repository has 6 quant files, from 31.1 GB (IQ2_XXS) to 47.9 GB (Q3_K_XS).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_XXS31.1 GB 36.2 GB
Too big Get file →
IQ2_XS34.6 GB 39.7 GB
Too big Get file →
Q2_K_S40.0 GB 45.1 GB
Too big Get file →
Q2_K43.2 GB 48.3 GB
Too big Get file →
IQ3_XXS45.9 GB 51.0 GB
Too big Get file →
Q3_K_XS47.9 GB 53.0 GB
Too big Get file →

What GPU do you need for miquella 120b?

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 Q2_K_S (40.0 GB file, about 45.1 GB in use).

Which miquella 120b quant should you choose?

Q2_K_S (40.0 GB) is the middle option in this repository. 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.

If memory is tight, the smallest file is IQ2_XXS at 31.1 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
Parameters118B
Layers137
Trained context32,764 tokens
KV cache (fp16)548 KB per token

Source

Quantized byalpindale
FamilyMiquella
Downloads (HF)63
Last updated2024-02-01

Frequently asked questions about miquella 120b

How much VRAM does miquella 120b need?

About 45.1 GB at Q2_K_S with an 8K context window. That is the 40.0 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 miquella 120b on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 36.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of miquella 120b GGUF?

It depends on the quant: IQ2_XXS is 31.1 GB, IQ2_XS is 34.6 GB, Q2_K_S is 40.0 GB, Q2_K is 43.2 GB, IQ3_XXS is 45.9 GB, Q3_K_XS is 47.9 GB.

What context length does miquella 120b support?

The GGUF header lists a trained context length of 32,764 tokens (32K). The KV cache grows with the context you use: about 4.6 GB at 8K and 18.4 GB at 32K.

Where can I download miquella 120b GGUF?

The files are on Hugging Face in the alpindale/miquella-120b-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 alpindale

All by alpindale →

More Miquella models

All Miquella →

Other large models

All large models →