Quantization made by Richard Erkhov.

Macaroni-v2-7b - GGUF

Model creator: https://huggingface.co/andrijdavid/
Original model: https://huggingface.co/andrijdavid/Macaroni-v2-7b/

Name	Quant method	Size
Macaroni-v2-7b.Q2_K.gguf	Q2_K	2.53GB
Macaroni-v2-7b.IQ3_XS.gguf	IQ3_XS	2.81GB
Macaroni-v2-7b.IQ3_S.gguf	IQ3_S	2.96GB
Macaroni-v2-7b.Q3_K_S.gguf	Q3_K_S	2.95GB
Macaroni-v2-7b.IQ3_M.gguf	IQ3_M	3.06GB
Macaroni-v2-7b.Q3_K.gguf	Q3_K	3.28GB
Macaroni-v2-7b.Q3_K_M.gguf	Q3_K_M	3.28GB
Macaroni-v2-7b.Q3_K_L.gguf	Q3_K_L	3.56GB
Macaroni-v2-7b.IQ4_XS.gguf	IQ4_XS	3.67GB
Macaroni-v2-7b.Q4_0.gguf	Q4_0	3.83GB
Macaroni-v2-7b.IQ4_NL.gguf	IQ4_NL	3.87GB
Macaroni-v2-7b.Q4_K_S.gguf	Q4_K_S	3.86GB
Macaroni-v2-7b.Q4_K.gguf	Q4_K	4.07GB
Macaroni-v2-7b.Q4_K_M.gguf	Q4_K_M	4.07GB
Macaroni-v2-7b.Q4_1.gguf	Q4_1	4.24GB
Macaroni-v2-7b.Q5_0.gguf	Q5_0	4.65GB
Macaroni-v2-7b.Q5_K_S.gguf	Q5_K_S	4.65GB
Macaroni-v2-7b.Q5_K.gguf	Q5_K	4.78GB
Macaroni-v2-7b.Q5_K_M.gguf	Q5_K_M	4.78GB
Macaroni-v2-7b.Q5_1.gguf	Q5_1	5.07GB
Macaroni-v2-7b.Q6_K.gguf	Q6_K	5.53GB
Macaroni-v2-7b.Q8_0.gguf	Q8_0	7.17GB

Original model description:

base_model: - flemmingmiguel/MBX-7B-v3 - mlabonne/OmniBeagle-7B - mistralai/Mistral-7B-v0.1 - vanillaOVO/supermario_v4 tags: - mergekit - merge license: apache-2.0 language: - en

Macaroni V2 7B

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using mistralai/Mistral-7B-v0.1 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: mistralai/Mistral-7B-v0.1
    # no parameters necessary for base model
  - model: flemmingmiguel/MBX-7B-v3
    parameters:
      density: 0.7
      weight: 0.5
  - model: vanillaOVO/supermario_v4
    parameters:
      density: 0.7
      weight: 0.3
  - model: mlabonne/OmniBeagle-7B
    parameters:
      density: 0.5
      weight: 0.6
merge_method: dare_ties
base_model: mistralai/Mistral-7B-v0.1
parameters:
  int8_mask: true
  normalize: true
dtype: float16