Quantization made by Richard Erkhov.

puffin-linear-coder-math-2 - GGUF

Model creator: https://huggingface.co/tuanpasg/
Original model: https://huggingface.co/tuanpasg/puffin-linear-coder-math-2/

Name	Quant method	Size
puffin-linear-coder-math-2.Q2_K.gguf	Q2_K	0.7GB
puffin-linear-coder-math-2.IQ3_XS.gguf	IQ3_XS	0.77GB
puffin-linear-coder-math-2.IQ3_S.gguf	IQ3_S	0.8GB
puffin-linear-coder-math-2.Q3_K_S.gguf	Q3_K_S	0.8GB
puffin-linear-coder-math-2.IQ3_M.gguf	IQ3_M	0.82GB
puffin-linear-coder-math-2.Q3_K.gguf	Q3_K	0.86GB
puffin-linear-coder-math-2.Q3_K_M.gguf	Q3_K_M	0.86GB
puffin-linear-coder-math-2.Q3_K_L.gguf	Q3_K_L	0.91GB
puffin-linear-coder-math-2.IQ4_XS.gguf	IQ4_XS	0.96GB
puffin-linear-coder-math-2.Q4_0.gguf	Q4_0	0.99GB
puffin-linear-coder-math-2.IQ4_NL.gguf	IQ4_NL	1.0GB
puffin-linear-coder-math-2.Q4_K_S.gguf	Q4_K_S	1.0GB
puffin-linear-coder-math-2.Q4_K.gguf	Q4_K	1.04GB
puffin-linear-coder-math-2.Q4_K_M.gguf	Q4_K_M	1.04GB
puffin-linear-coder-math-2.Q4_1.gguf	Q4_1	1.08GB
puffin-linear-coder-math-2.Q5_0.gguf	Q5_0	1.17GB
puffin-linear-coder-math-2.Q5_K_S.gguf	Q5_K_S	1.17GB
puffin-linear-coder-math-2.Q5_K.gguf	Q5_K	1.2GB
puffin-linear-coder-math-2.Q5_K_M.gguf	Q5_K_M	1.2GB
puffin-linear-coder-math-2.Q5_1.gguf	Q5_1	1.26GB
puffin-linear-coder-math-2.Q6_K.gguf	Q6_K	1.36GB
puffin-linear-coder-math-2.Q8_0.gguf	Q8_0	1.76GB

Original model description:

base_model:

Qwen/Qwen2.5-1.5B
Qwen/Qwen2.5-Coder-1.5B
Qwen/Qwen2.5-Math-1.5B 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 TIES merge method using Qwen/Qwen2.5-1.5B 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: Qwen/Qwen2.5-Coder-1.5B
    parameters:
      density: 0.5
      weight: 0.9
  - model: Qwen/Qwen2.5-Math-1.5B
    parameters:
      density: 0.5
      weight: 0.1
merge_method: ties
base_model: Qwen/Qwen2.5-1.5B
parameters:
  normalize: true
  int8_mask: true
tokenizer_source: Qwen/Qwen2.5-Coder-1.5B
dtype: bfloat16