knifeayumu/Cydonia-v4-MS3.2-Magnum-Diamond-24B-GGUF

Llamacpp Quantizations of knifeayumu/Cydonia-v4-MS3.2-Magnum-Diamond-24B

Using llama.cpp release b5966 for quantization.

Original model: knifeayumu/Cydonia-v4-MS3.2-Magnum-Diamond-24B

Quant Types:

Filename	Quant type	File Size
Cydonia-v4-MS3.2-Magnum-Diamond-24B-F16.gguf	F16	47.15 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q8_0.gguf	Q8_0	25.05 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q6_K.gguf	Q6_K	19.35 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q5_K_M.gguf	Q5_K_M	16.76 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q5_K_S.gguf	Q5_K_S	16.30 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q4_K_M.gguf	Q4_K_M	14.33 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q4_K_S.gguf	Q4_K_S	13.55 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q3_K_L.gguf	Q3_K_L	12.40 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q3_K_M.gguf	Q3_K_M	11.47 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q3_K_S.gguf	Q3_K_S	10.40 GB
Cydonia-v4-MS3.2-Magnum-Diamond-24B-Q2_K.gguf	Q2_K	8.89 GB

---

Cydonia-v4-MS3.2-Magnum-Diamond-24B

Recipe based on knifeayumu/Cydonia-v1.2-Magnum-v4-22B because the model Doctor-Shotgun/MS3.2-24B-Magnum-Diamond is still too horny and verbose.

The PNG file above includes workflow for FLUX Kontext Dev with ComfyUI utilising pollockjj/ComfyUI-MultiGPU nodes and two input images without stitching.

Merge Details

This is a merge of pre-trained language models created using mergekit.

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

• TheDrummer/Cydonia-24B-v4
• Doctor-Shotgun/MS3.2-24B-Magnum-Diamond

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: TheDrummer/Cydonia-24B-v4
  - model: Doctor-Shotgun/MS3.2-24B-Magnum-Diamond
merge_method: slerp
base_model: TheDrummer/Cydonia-24B-v4
parameters:
  t: [0.1, 0.3, 0.6, 0.3, 0.1]
dtype: bfloat16