HU

HuginnV5.5 12.6B

HuginnvGGUF by LoneStrikerUpdated 2024-01-28
Download on Hugging Face
Parameters12.9B
Context32K
Layers58
Architecturellama
Smallest quant5.6 GB
Quick answer

HuginnV5.5 12.6B needs about 10.2 GB of memory at Q4_K_M (a 7.8 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 9 quants, from 5.6 GB (Q3_K_S) to 13.7 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 7.8 GB.

HuginnV5.5 12.6B is a 12.9B-parameter Huginnv model in GGUF format, quantized by LoneStriker. Its trained context length is 32K tokens. The repository has 9 quant files, from 5.6 GB (Q3_K_S) to 13.7 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S5.6 GB 8.0 GB
Too big Get file →
Q3_K_M6.2 GB 8.7 GB
Too big Get file →
Q3_K_L6.8 GB 9.2 GB
Too big Get file →
Q4_K_S7.4 GB 9.8 GB
Too big Get file →
Q4_K_M7.8 GB 10.2 GB
Too big Get file →
Q5_K_S8.9 GB 11.3 GB
Too big Get file →
Q5_K_M9.1 GB 11.6 GB
Too big Get file →
Q6_K10.6 GB 13.0 GB
Too big Get file →
Q8_013.7 GB 16.2 GB
Too big Get file →

What GPU do you need for HuginnV5.5 12.6B?

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: the best quant that fits is Q3_K_S (5.6 GB file, about 8.0 GB in use).
  • 12 GB VRAM: the best quant that fits is Q5_K_M (9.1 GB file, about 11.6 GB in use).
  • 16 GB VRAM: the best quant that fits is Q6_K (10.6 GB file, about 13.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (13.7 GB file, about 16.2 GB in use).

Which HuginnV5.5 12.6B quant should you choose?

Q4_K_M is the usual starting point for HuginnV5.5 12.6B: it balances file size and quality well. Here it is a 7.8 GB file.

The Q8_0 file is 13.7 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q3_K_S at 5.6 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters12.9B
Layers58
Trained context32,768 tokens
KV cache (fp16)232 KB per token

Source

Quantized byLoneStriker
FamilyHuginnv
Licensecc-by-4.0
Downloads (HF)14
Last updated2024-01-28

Frequently asked questions about HuginnV5.5 12.6B

How much VRAM does HuginnV5.5 12.6B need?

About 10.2 GB at Q4_K_M with an 8K context window. That is the 7.8 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 HuginnV5.5 12.6B on an 8 GB GPU?

Yes, with Q3_K_S: the 5.6 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of HuginnV5.5 12.6B GGUF?

It depends on the quant: Q3_K_S is 5.6 GB, Q3_K_M is 6.2 GB, Q3_K_L is 6.8 GB, Q4_K_S is 7.4 GB, Q4_K_M is 7.8 GB, Q5_K_S is 8.9 GB and 3 more.

What context length does HuginnV5.5 12.6B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.9 GB at 8K and 7.8 GB at 32K.

Where can I download HuginnV5.5 12.6B GGUF?

The files are on Hugging Face in the LoneStriker/HuginnV5.5-12.6B-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 HuginnV5.5 12.6B use?

Hugging Face lists the license as cc-by-4.0. 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 LoneStriker

All by LoneStriker →

More Huginnv models

All Huginnv →

Other medium models

All medium models →