Downloads · 30 days
85
6% of all-time downloads
RichardErkhov/mergekit-community_-_HX-Mistral-3B_v0.1-gguf
mergekit-community_-_HX-Mistral-3B_v0.1-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
85
6% of all-time downloads
All-time downloads
1.4K
Public
Repo size
37.9 GB
Likes
0
Public
Click a slice to open those files.
.gguf44.7 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
HX-Mistral-3B_v0.1 - GGUF
| Name | Quant method | Size |
|---|---|---|
| HX-Mistral-3B_v0.1.Q2_K.gguf | Q2_K | 1.34GB |
| HX-Mistral-3B_v0.1.Q3_K_S.gguf | Q3_K_S | 1.55GB |
| HX-Mistral-3B_v0.1.Q3_K.gguf | Q3_K | 1.72GB |
| HX-Mistral-3B_v0.1.Q3_K_M.gguf | Q3_K_M | 1.72GB |
| HX-Mistral-3B_v0.1.Q3_K_L.gguf | Q3_K_L | 1.86GB |
| HX-Mistral-3B_v0.1.IQ4_XS.gguf | IQ4_XS | 1.92GB |
| HX-Mistral-3B_v0.1.Q4_0.gguf | Q4_0 | 2.0GB |
| HX-Mistral-3B_v0.1.IQ4_NL.gguf | IQ4_NL | 2.02GB |
| HX-Mistral-3B_v0.1.Q4_K_S.gguf | Q4_K_S | 2.01GB |
| HX-Mistral-3B_v0.1.Q4_K.gguf | Q4_K | 2.12GB |
| HX-Mistral-3B_v0.1.Q4_K_M.gguf | Q4_K_M | 2.12GB |
| HX-Mistral-3B_v0.1.Q4_1.gguf | Q4_1 | 2.21GB |
| HX-Mistral-3B_v0.1.Q5_0.gguf | Q5_0 | 2.42GB |
| HX-Mistral-3B_v0.1.Q5_K_S.gguf | Q5_K_S | 2.42GB |
| HX-Mistral-3B_v0.1.Q5_K.gguf | Q5_K | 2.48GB |
| HX-Mistral-3B_v0.1.Q5_K_M.gguf | Q5_K_M | 2.48GB |
| HX-Mistral-3B_v0.1.Q5_1.gguf | Q5_1 | 2.63GB |
| HX-Mistral-3B_v0.1.Q6_K.gguf | Q6_K | 2.87GB |
| HX-Mistral-3B_v0.1.Q8_0.gguf | Q8_0 | 3.71GB |
base_model:
This is a merge of pre-trained language models created using mergekit.
This model was merged using the linear merge method.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
dtype: float16
merge_method: linear
slices:
- sources:
- layer_range: [0, 16] # Assuming the first half of the model is more general and can be reduced more
model: mistralai/Mistral-7B-Instruct-v0.2
parameters:
weight: 0.5 # Reduce the weight of the first half to make room for the second half
- layer_range: [16, 32] # Assuming the second half of the model is more specialized and can be reduced less
model: mistralai/Mistral-7B-Instruct-v0.2
parameters:
weight: 0.5 # Maintain the weight of the second half