ME

MedMitsu

MedmitsuGGUF by HimitsuiUpdated 2024-01-20
Download on Hugging Face
Parameters10.7B
Context4K
Layers48
Architecturellama
Smallest quant6.1 GB
Quick answer

MedMitsu needs about 9.7 GB of memory at Q5_K_M (a 7.6 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 6 quants, from 6.1 GB (Q4_K_S) to 8.8 GB (Q6_K).

MedMitsu is a 10.7B-parameter Medmitsu model in GGUF format, quantized by Himitsui. Its trained context length is 4K tokens. The repository has 6 quant files, from 6.1 GB (Q4_K_S) to 8.8 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_S6.1 GB 7.4 GB
Tight Get file →
Q4_K_S6.1 GB 7.4 GB
Tight Get file →
Q5_K_M7.6 GB 8.9 GB
Too big Get file →
Q5_K_M7.6 GB 8.9 GB
Too big Get file →
Q6_K8.8 GB 10.1 GB
Too big Get file →
Q6_K8.8 GB 10.1 GB
Too big Get file →

What GPU do you need for MedMitsu?

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 (8.8 GB file, about 10.9 GB in use).

Which MedMitsu quant should you choose?

Q5_K_M (7.6 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

If memory is tight, the smallest file is Q4_K_S at 6.1 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters10.7B
Layers48
Trained context4,096 tokens
KV cache (fp16)192 KB per token

Source

Quantized byHimitsui
FamilyMedmitsu
Downloads (HF)123
Last updated2024-01-20

Frequently asked questions about MedMitsu

How much VRAM does MedMitsu need?

About 9.7 GB at Q5_K_M with an 8K context window. That is the 7.6 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 MedMitsu on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 8.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MedMitsu GGUF?

It depends on the quant: Q4_K_S is 6.1 GB, Q4_K_S is 6.1 GB, Q5_K_M is 7.6 GB, Q5_K_M is 7.6 GB, Q6_K is 8.8 GB, Q6_K is 8.8 GB.

What context length does MedMitsu support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 1.6 GB at 8K and 0.8 GB at 4K.

Where can I download MedMitsu GGUF?

The files are on Hugging Face in the Himitsui/MedMitsu-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 MedMitsu use?

Hugging Face lists the license as cc-by-nc-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 Himitsui

All by Himitsui →

Other medium models

All medium models →