RichardErkhov/TroyDoesAI_-_Llama-3.1-13B-Instruct-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Llama-3.1-13B-Instruct - GGUF

• Model creator: https://huggingface.co/TroyDoesAI/
• Original model: https://huggingface.co/TroyDoesAI/Llama-3.1-13B-Instruct/

Name	Quant method	Size
Llama-3.1-13B-Instruct.Q2_K.gguf	Q2_K	4.75GB
Llama-3.1-13B-Instruct.Q3_K_S.gguf	Q3_K_S	5.51GB
Llama-3.1-13B-Instruct.Q3_K.gguf	Q3_K	6.08GB
Llama-3.1-13B-Instruct.Q3_K_M.gguf	Q3_K_M	6.08GB
Llama-3.1-13B-Instruct.Q3_K_L.gguf	Q3_K_L	6.58GB
Llama-3.1-13B-Instruct.IQ4_XS.gguf	IQ4_XS	6.81GB
Llama-3.1-13B-Instruct.Q4_0.gguf	Q4_0	7.08GB
Llama-3.1-13B-Instruct.IQ4_NL.gguf	IQ4_NL	7.16GB
Llama-3.1-13B-Instruct.Q4_K_S.gguf	Q4_K_S	7.13GB
Llama-3.1-13B-Instruct.Q4_K.gguf	Q4_K	7.51GB
Llama-3.1-13B-Instruct.Q4_K_M.gguf	Q4_K_M	7.51GB
Llama-3.1-13B-Instruct.Q4_1.gguf	Q4_1	7.83GB
Llama-3.1-13B-Instruct.Q5_0.gguf	Q5_0	8.57GB
Llama-3.1-13B-Instruct.Q5_K_S.gguf	Q5_K_S	8.57GB
Llama-3.1-13B-Instruct.Q5_K.gguf	Q5_K	8.78GB
Llama-3.1-13B-Instruct.Q5_K_M.gguf	Q5_K_M	8.78GB
Llama-3.1-13B-Instruct.Q5_1.gguf	Q5_1	9.31GB
Llama-3.1-13B-Instruct.Q6_K.gguf	Q6_K	10.14GB
Llama-3.1-13B-Instruct.Q8_0.gguf	Q8_0	13.13GB

Original model description:

base_model:

• TroyDoesAI/Llama-3.1-8B-Instruct library_name: transformers tags:
• mergekit
• merge

---

Llama-3.1-13B-Instruct

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

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

• TroyDoesAI/Llama-3.1-8B-Instruct

Configuration

The following YAML configuration was used to produce this model:

#### BEST CONFIGURATION ####
slices:
- sources:
  - layer_range: [0, 8]
    model: TroyDoesAI/Llama-3.1-8B-Instruct
- sources:
  - layer_range: [4, 12]
    model: TroyDoesAI/Llama-3.1-8B-Instruct
- sources:
  - layer_range: [8, 16]
    model: TroyDoesAI/Llama-3.1-8B-Instruct
- sources:
  - layer_range: [12, 20]
    model: TroyDoesAI/Llama-3.1-8B-Instruct
- sources:
  - layer_range: [16, 24]
    model: TroyDoesAI/Llama-3.1-8B-Instruct
- sources:
  - layer_range: [20, 28]
    model: TroyDoesAI/Llama-3.1-8B-Instruct
- sources:
  - layer_range: [24, 32]
    model: TroyDoesAI/Llama-3.1-8B-Instruct
merge_method: passthrough
dtype: bfloat16