MA

Marx-3B-V3_safetensors

MarxGGUF by RachidARUpdated 2023-10-24
Download on Hugging Face
Parameters2.8B
Context4K
Layers32
Architecturestablelm
Smallest quant1.4 GB
Quick answer

Marx-3B-V3_safetensors needs about 4.9 GB of memory at Q4_K_M (a 1.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 6 quants, from 1.4 GB (Q3_K_M) to 5.6 GB (F16).

Marx-3B-V3_safetensors is a 2.8B-parameter Marx model in GGUF format, quantized by RachidAR. Its trained context length is 4K tokens. The repository has 6 quant files, from 1.4 GB (Q3_K_M) to 5.6 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_M1.4 GB 3.2 GB
Fits Get file →
Q4_K_M1.7 GB 3.6 GB
Fits Get file →
Q4_11.8 GB 3.6 GB
Fits Get file →
Q5_K_M2.0 GB 3.8 GB
Fits Get file →
Q6_K2.3 GB 4.1 GB
Fits Get file →
F165.6 GB 7.4 GB
Tight Get file →

What GPU do you need for Marx-3B-V3_safetensors?

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 Q6_K (2.3 GB file, about 5.5 GB in use).
  • 8 GB VRAM: the best quant that fits is Q6_K (2.3 GB file, about 5.5 GB in use).
  • 12 GB VRAM: the best quant that fits is F16 (5.6 GB file, about 8.8 GB in use).

Which Marx-3B-V3_safetensors quant should you choose?

Q4_K_M is the usual starting point for Marx-3B-V3_safetensors: it balances file size and quality well. Here it is a 1.7 GB file.

If memory is tight, the smallest file is Q3_K_M at 1.4 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

Architecturestablelm
Parameters2.8B
Layers32
Trained context4,096 tokens
KV cache (fp16)320 KB per token

Source

Quantized byRachidAR
FamilyMarx
Downloads (HF)457
Last updated2023-10-24

Frequently asked questions about Marx-3B-V3_safetensors

How much VRAM does Marx-3B-V3_safetensors need?

About 4.9 GB at Q4_K_M with an 8K context window. That is the 1.7 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 Marx-3B-V3_safetensors on an 8 GB GPU?

Yes, with Q6_K: the 2.3 GB file needs about 5.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Marx-3B-V3_safetensors GGUF?

It depends on the quant: Q3_K_M is 1.4 GB, Q4_K_M is 1.7 GB, Q4_1 is 1.8 GB, Q5_K_M is 2.0 GB, Q6_K is 2.3 GB, F16 is 5.6 GB.

What context length does Marx-3B-V3_safetensors support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 2.7 GB at 8K and 1.3 GB at 4K.

Where can I download Marx-3B-V3_safetensors GGUF?

The files are on Hugging Face in the RachidAR/Marx-3b-v3-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 Marx-3B-V3_safetensors use?

Hugging Face lists the license as cc-by-sa-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 RachidAR

All by RachidAR →

More Marx models

All Marx →

Other tiny models

All tiny models →