Quantization made by Richard Erkhov.

Github

Discord

Request more models

dare_saba_1 - GGUF

• Model creator: https://huggingface.co/Sakalti/
• Original model: https://huggingface.co/Sakalti/dare_saba_1/

Name	Quant method	Size
dare_saba_1.Q2_K.gguf	Q2_K	1.08GB
dare_saba_1.IQ3_XS.gguf	IQ3_XS	1.2GB
dare_saba_1.IQ3_S.gguf	IQ3_S	1.25GB
dare_saba_1.Q3_K_S.gguf	Q3_K_S	1.25GB
dare_saba_1.IQ3_M.gguf	IQ3_M	1.28GB
dare_saba_1.Q3_K.gguf	Q3_K	1.36GB
dare_saba_1.Q3_K_M.gguf	Q3_K_M	1.36GB
dare_saba_1.Q3_K_L.gguf	Q3_K_L	1.46GB
dare_saba_1.IQ4_XS.gguf	IQ4_XS	1.52GB
dare_saba_1.Q4_0.gguf	Q4_0	1.58GB
dare_saba_1.IQ4_NL.gguf	IQ4_NL	1.59GB
dare_saba_1.Q4_K_S.gguf	Q4_K_S	1.59GB
dare_saba_1.Q4_K.gguf	Q4_K	1.67GB
dare_saba_1.Q4_K_M.gguf	Q4_K_M	1.67GB
dare_saba_1.Q4_1.gguf	Q4_1	1.74GB
dare_saba_1.Q5_0.gguf	Q5_0	1.89GB
dare_saba_1.Q5_K_S.gguf	Q5_K_S	1.89GB
dare_saba_1.Q5_K.gguf	Q5_K	1.94GB
dare_saba_1.Q5_K_M.gguf	Q5_K_M	1.94GB
dare_saba_1.Q5_1.gguf	Q5_1	2.05GB
dare_saba_1.Q6_K.gguf	Q6_K	2.22GB
dare_saba_1.Q8_0.gguf	Q8_0	2.88GB

Original model description:

base_model:

• win10/Qwen2.5-2B-Instruct 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 win10/Qwen2.5-2B-Instruct 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: win10/Qwen2.5-2B-Instruct
    parameters:
      density: 0.5
      weight: 0.4
merge_method: dare_ties
base_model: win10/Qwen2.5-2B-Instruct
parameters:
  density: 0.5
  weight: 0.6
  int8_mask: true
dtype: float16