MA

madlad400 3b mt

MadladGGUF by hs-hfUpdated 2024-10-28
Download on Hugging Face
Parameters2.94B
Context512
Layers32
Architecturet5
Smallest quant1.7 GB
Quick answer

madlad400 3b mt needs about 4.5 GB of memory at Q4_K_M (a 1.9 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 7 quants, from 1.7 GB (Q4_K_S) to 5.9 GB.

madlad400 3b mt is a 2.94B-parameter Madlad model in GGUF format, quantized by hs-hf. Its trained context length is 512 tokens. The repository has 7 quant files, from 1.7 GB (Q4_K_S) to 5.9 GB.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_S1.7 GB 2.4 GB
Fits Get file →
Q4_K_M1.9 GB 2.5 GB
Fits Get file →
Q5_K_S2.1 GB 2.7 GB
Fits Get file →
Q5_K_M2.1 GB 2.8 GB
Fits Get file →
Q6_K2.4 GB 3.1 GB
Fits Get file →
Q8_03.1 GB 3.8 GB
Fits Get file →
—5.9 GB 6.5 GB
Fits Get file →

What GPU do you need for madlad400 3b mt?

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: the best quant that fits is Q8_0 (3.1 GB file, about 5.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (3.1 GB file, about 5.8 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (5.9 GB file, about 8.5 GB in use).

Which madlad400 3b mt quant should you choose?

Q4_K_M is the usual starting point for madlad400 3b mt: it balances file size and quality well. Here it is a 1.9 GB file.

The Q8_0 file is 3.1 GB, about 1.7 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 Q4_K_S at 1.7 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

Architecturet5
Parameters2.94B
Layers32
Trained context512 tokens
KV cache (fp16)256 KB per token

Source

Quantized byhs-hf
FamilyMadlad
Licenseapache-2.0
Downloads (HF)332
Last updated2024-10-28

Frequently asked questions about madlad400 3b mt

How much VRAM does madlad400 3b mt need?

About 4.5 GB at Q4_K_M with an 8K context window. That is the 1.9 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 madlad400 3b mt on an 8 GB GPU?

Yes, with Q8_0: the 3.1 GB file needs about 5.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of madlad400 3b mt GGUF?

It depends on the quant: Q4_K_S is 1.7 GB, Q4_K_M is 1.9 GB, Q5_K_S is 2.1 GB, Q5_K_M is 2.1 GB, Q6_K is 2.4 GB, Q8_0 is 3.1 GB and 1 more.

What context length does madlad400 3b mt support?

The GGUF header lists a trained context length of 512 tokens (512). The KV cache grows with the context you use: about 2.1 GB at 8K and 0.1 GB at 512.

Where can I download madlad400 3b mt GGUF?

The files are on Hugging Face in the hs-hf/madlad400-3b-mt-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 madlad400 3b mt use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Other GGUFs of madlad400-3b-mt

Similar models

Browse all →

More from hs-hf

All by hs-hf →

More Madlad models

All Madlad →

Other tiny models

All tiny models →