TE

Test Frankenstein

TestGGUF by ND911Updated 2024-11-29
Download on Hugging Face
Parameters11.9B
Context—
Layers—
Architectureflux
Smallest quant6.9 GB
Quick answer

Test Frankenstein needs about 14.7 GB of memory at Q8_0 (a 12.7 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 5 quants, from 6.9 GB (Q4_K) to 23.9 GB (BF16).

Test Frankenstein – a distilled flux model for image generation

Test Frankenstein is a 11.9‑billion parameter flux model. It is designed for generating images in 12 steps with a refiner. The model supports flux loras and should work with sdxl loras.

Good at

  • Fast image generation in 12 steps
  • High quality output with refiner
  • Compatibility with flux and sdxl loras

Limits

  • No vision support beyond image generation
  • Requires clips, vae and CFG
  • FP16 not supported – use FP8

Best for

  • Rapid prototyping of image generation
  • Testing flux loras
  • Generating high‑quality images with a refiner

Who it’s for. Developers and researchers who need a lightweight flux model for image generation. The model is suitable for testing and experimentation with loras and refiner workflows.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K6.9 GB 8.2 GB
Too big Get file →
Q6_K9.8 GB 11.5 GB
Too big Get file →
Q8_012.7 GB 14.7 GB
Too big Get file →
Q8_012.7 GB 14.7 GB
Too big Get file →
BF1623.9 GB 27.3 GB
Too big Get file →

For 8 GB VRAM, use Q4_K (file size 6.91 GB, estimated memory 8.2 GB). For 12 GB VRAM, Q6_K (9.83 GB, 11.5 GB). For 20 GB VRAM, Q8_0 (12.69 GB, 14.7 GB). BF16 requires 23.9 GB and 27.3 GB memory.

What GPU do you need for Test Frankenstein?

These figures add the file size, an fp16 KV cache at an 8K context window 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 (9.8 GB file, about 11.5 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (12.7 GB file, about 14.7 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (12.7 GB file, about 14.7 GB in use).
  • 32 GB VRAM: the best quant that fits is BF16 (23.9 GB file, about 27.3 GB in use).

Which Test Frankenstein quant should you choose?

Q8_0 (12.7 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

If memory is tight, the smallest file is Q4_K at 6.9 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

Architectureflux
Parameters11.9B

Source

Quantized byND911
FamilyTest
Licenseapache-2.0
Downloads (HF)17,186
Last updated2024-11-29

Frequently asked questions about Test Frankenstein

How much VRAM does Test Frankenstein need?

About 14.7 GB at Q8_0 with an 8K context window. That is the 12.7 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 Test Frankenstein 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 Test Frankenstein GGUF?

It depends on the quant: Q4_K is 6.9 GB, Q6_K is 9.8 GB, Q8_0 is 12.7 GB, Q8_0 is 12.7 GB, BF16 is 23.9 GB.

Where can I download Test Frankenstein GGUF?

The files are on Hugging Face in the ND911/Test_Frankenstein repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Test Frankenstein 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.

Can I run Q4_K on 8 GB VRAM?

Yes, Q4_K uses 8.2 GB estimated memory at 8k context, which fits within 8 GB VRAM with some headroom.

Does the model support FP16?

No, FP16 is not supported. Use FP8 to avoid high VRAM usage.

What additional files are needed?

Clips, VAE and CFG are required for proper operation.

Similar models

Browse all →

More from ND911

All by ND911 →

More Test models

All Test →

Other medium models

All medium models →