Quantization made by Richard Erkhov.

llama3-8B-Special-Dark-v3.1.2c - GGUF

Model creator: https://huggingface.co/MrRobotoAI/
Original model: https://huggingface.co/MrRobotoAI/llama3-8B-Special-Dark-v3.1.2c/

Name	Quant method	Size
llama3-8B-Special-Dark-v3.1.2c.Q2_K.gguf	Q2_K	2.96GB
llama3-8B-Special-Dark-v3.1.2c.IQ3_XS.gguf	IQ3_XS	3.28GB
llama3-8B-Special-Dark-v3.1.2c.IQ3_S.gguf	IQ3_S	3.43GB
llama3-8B-Special-Dark-v3.1.2c.Q3_K_S.gguf	Q3_K_S	3.41GB
llama3-8B-Special-Dark-v3.1.2c.IQ3_M.gguf	IQ3_M	2.76GB
llama3-8B-Special-Dark-v3.1.2c.Q3_K.gguf	Q3_K	3.74GB
llama3-8B-Special-Dark-v3.1.2c.Q3_K_M.gguf	Q3_K_M	2.1GB
llama3-8B-Special-Dark-v3.1.2c.Q3_K_L.gguf	Q3_K_L	3.6GB
llama3-8B-Special-Dark-v3.1.2c.IQ4_XS.gguf	IQ4_XS	1.17GB
llama3-8B-Special-Dark-v3.1.2c.Q4_0.gguf	Q4_0	4.34GB
llama3-8B-Special-Dark-v3.1.2c.IQ4_NL.gguf	IQ4_NL	4.38GB
llama3-8B-Special-Dark-v3.1.2c.Q4_K_S.gguf	Q4_K_S	4.37GB
llama3-8B-Special-Dark-v3.1.2c.Q4_K.gguf	Q4_K	4.58GB
llama3-8B-Special-Dark-v3.1.2c.Q4_K_M.gguf	Q4_K_M	4.58GB
llama3-8B-Special-Dark-v3.1.2c.Q4_1.gguf	Q4_1	4.78GB
llama3-8B-Special-Dark-v3.1.2c.Q5_0.gguf	Q5_0	5.21GB
llama3-8B-Special-Dark-v3.1.2c.Q5_K_S.gguf	Q5_K_S	5.21GB
llama3-8B-Special-Dark-v3.1.2c.Q5_K.gguf	Q5_K	5.34GB
llama3-8B-Special-Dark-v3.1.2c.Q5_K_M.gguf	Q5_K_M	5.34GB
llama3-8B-Special-Dark-v3.1.2c.Q5_1.gguf	Q5_1	5.65GB
llama3-8B-Special-Dark-v3.1.2c.Q6_K.gguf	Q6_K	6.14GB
llama3-8B-Special-Dark-v3.1.2c.Q8_0.gguf	Q8_0	7.95GB

Original model description:

base_model:

MrRobotoAI/llama3-8B-Special-Dark-RP2
MrRobotoAI/llama3-8B-Special-Dark-v3.1.2b
MrRobotoAI/Loki-base library_name: transformers tags:
mergekit
merge

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 MrRobotoAI/llama3-8B-Special-Dark-RP2 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: MrRobotoAI/Loki-base
    parameters:
      weight: 0.4
      density: 0.9
  - model: MrRobotoAI/llama3-8B-Special-Dark-v3.1.2b
    parameters:
      weight: 0.4
      density: 0.9
  - model: MrRobotoAI/llama3-8B-Special-Dark-RP2
    parameters:
      weight: 0.2
      density: 0.9

merge_method: dare_ties
base_model: MrRobotoAI/llama3-8B-Special-Dark-RP2
parameters:
  normalize: true
  int8_mask: true
dtype: float16