Name	Quant method	Size
Ultra-Instruct-12B.Q2_K.gguf	Q2_K	4.46GB
Ultra-Instruct-12B.Q3_K_S.gguf	Q3_K_S	5.15GB
Ultra-Instruct-12B.Q3_K.gguf	Q3_K	5.67GB
Ultra-Instruct-12B.Q3_K_M.gguf	Q3_K_M	5.67GB
Ultra-Instruct-12B.Q3_K_L.gguf	Q3_K_L	6.11GB
Ultra-Instruct-12B.IQ4_XS.gguf	IQ4_XS	6.33GB
Ultra-Instruct-12B.Q4_0.gguf	Q4_0	6.59GB
Ultra-Instruct-12B.IQ4_NL.gguf	IQ4_NL	6.65GB
Ultra-Instruct-12B.Q4_K_S.gguf	Q4_K_S	6.63GB
Ultra-Instruct-12B.Q4_K.gguf	Q4_K	6.96GB
Ultra-Instruct-12B.Q4_K_M.gguf	Q4_K_M	6.96GB
Ultra-Instruct-12B.Q4_1.gguf	Q4_1	7.26GB
Ultra-Instruct-12B.Q5_0.gguf	Q5_0	7.93GB
Ultra-Instruct-12B.Q5_K_S.gguf	Q5_K_S	7.93GB
Ultra-Instruct-12B.Q5_K.gguf	Q5_K	8.13GB
Ultra-Instruct-12B.Q5_K_M.gguf	Q5_K_M	8.13GB
Ultra-Instruct-12B.Q5_1.gguf	Q5_1	8.61GB
Ultra-Instruct-12B.Q6_K.gguf	Q6_K	9.37GB
Ultra-Instruct-12B.Q8_0.gguf	Q8_0	12.13GB

Original model description:

base_model:

Pyroserenus/Orthrus-12b-v0.8
nbeerbower/mistral-nemo-gutenberg-12B-v2
mergekit-community/Deutscher-Pantheon-12B
IntervitensInc/Mistral-Nemo-Base-2407-chatml library_name: transformers tags:
mergekit
merge license: cc-by-nc-4.0

merge

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 IntervitensInc/Mistral-Nemo-Base-2407-chatml as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

yaml

models:
  - model: mergekit-community/Deutscher-Pantheon-12B
    parameters:
      weight: 0.3
      density: 0.5
  - model: nbeerbower/mistral-nemo-gutenberg-12B-v2
    parameters:
      weight: 0.3
      density: 0.5
  - model: Pyroserenus/Orthrus-12b-v0.8
    parameters:
      weight: 0.6
      density: 0.5
merge_method: dare_ties
base_model: IntervitensInc/Mistral-Nemo-Base-2407-chatml
dtype: bfloat16
name: Ultra-Instruct-12B

Not responsible for what you do with it. Use with caution. WARNING: UNCENSORED. SMART.

Use ChatML. Note: It appears it has trouble stopping. If you value extremly long replies, this might be the model for you.

RichardErkhov/Dampfinchen_-_Ultra-Instruct-12B-gguf