MI

MiniMax H3 Pruned Ref Delta Fused

MinimaxGGUF by AbirayUpdated 2026-09-16
Download on Hugging Face
Parameters20.1B
Context—
Layers—
Architecturewan
Smallest quant8.9 GB
Quick answer

MiniMax H3 Pruned Ref Delta Fused needs about 13.5 GB of memory at Q4_K_M (a 11.6 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 7 quants, from 8.9 GB (Q3_K_M) to 21.6 GB (Q8_0).

MiniMax H3 Pruned Ref Delta Fused – 20.1B parameter diffusion model

MiniMax H3 Pruned Ref Delta Fused is a 20.1B parameter diffusion transformer. It is available in GGUF format with several quantizations. The model is designed for reference-conditioned image generation on consumer GPUs.

Good at

  • Reference-conditioned diffusion
  • High visual quality
  • Motion dynamics

Limits

  • No text or audio support
  • Requires 16–24 GB VRAM for full quality
  • License is other

Best for

  • Local reference-based image synthesis
  • Video generation with reference frames
  • Fine-tuning on custom datasets

Who it’s for. Users who need a high‑quality diffusion model that can run locally on consumer GPUs. It is suitable for artists and developers working with reference-conditioned image and video generation.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_M8.9 GB 10.5 GB
Too big Get file →
Q4_K_M11.6 GB 13.5 GB
Too big Get file →
Q4_K_S11.6 GB 13.5 GB
Too big Get file →
Q5_K_M14.1 GB 16.3 GB
Too big Get file →
Q5_K_S14.1 GB 16.3 GB
Too big Get file →
Q6_K16.7 GB 19.2 GB
Too big Get file →
Q8_021.6 GB 24.7 GB
Too big Get file →

For 8 GB VRAM use Q3_K_M (8.9 GB file, 10.5 GB memory at 8k context). For 12 GB VRAM use Q4_K_M (11.56 GB file, 13.5 GB memory). For 16 GB VRAM use Q5_K_M (14.07 GB file, 16.3 GB memory). Memory estimates are approximate.

What GPU do you need for MiniMax H3 Pruned Ref Delta Fused?

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 Q3_K_M (8.9 GB file, about 10.5 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_M (11.6 GB file, about 13.5 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (16.7 GB file, about 19.2 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (21.6 GB file, about 24.7 GB in use).

Which MiniMax H3 Pruned Ref Delta Fused quant should you choose?

Q4_K_M is the usual starting point for MiniMax H3 Pruned Ref Delta Fused: it balances file size and quality well. Here it is a 11.6 GB file.

The Q8_0 file is 21.6 GB, about 1.9 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_M at 8.9 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

Architecturewan
Parameters20.1B

Source

Quantized byAbiray
FamilyMinimax
Licenseother
Downloads (HF)8,997
Last updated2026-09-16

Frequently asked questions about MiniMax H3 Pruned Ref Delta Fused

How much VRAM does MiniMax H3 Pruned Ref Delta Fused need?

About 13.5 GB at Q4_K_M with an 8K context window. That is the 11.6 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 MiniMax H3 Pruned Ref Delta Fused on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 10.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MiniMax H3 Pruned Ref Delta Fused GGUF?

It depends on the quant: Q3_K_M is 8.9 GB, Q4_K_M is 11.6 GB, Q4_K_S is 11.6 GB, Q5_K_M is 14.1 GB, Q5_K_S is 14.1 GB, Q6_K is 16.7 GB and 1 more.

Where can I download MiniMax H3 Pruned Ref Delta Fused GGUF?

The files are on Hugging Face in the Abiray/MiniMax-H3-Pruned-Ref-Delta-Fused-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 MiniMax H3 Pruned Ref Delta Fused use?

Hugging Face lists the license as other. 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 this model on 8 GB VRAM?

Use the Q3_K_M quantization; it requires about 10.5 GB memory at 8k context, which is close to 8 GB VRAM but may need swapping.

Does this model support text prompts?

No, the model only contains the diffusion transformer; it does not include a text encoder or tokenizer.

What is the context length of this model?

The trained context tokens are not specified; context length is not provided in the facts.

Similar models

Browse all →

More from Abiray

All by Abiray →

More Minimax models

All Minimax →

Other medium models

All medium models →