Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
Paper • 2311.03099 • Published • 38
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Quantization made by Richard Erkhov.
BioMistral-MedMNX - GGUF
- Model creator: https://huggingface.co/BioMistral/
- Original model: https://huggingface.co/BioMistral/BioMistral-MedMNX/
| Name | Quant method | Size |
|---|---|---|
| BioMistral-MedMNX.Q2_K.gguf | Q2_K | 2.53GB |
| BioMistral-MedMNX.IQ3_XS.gguf | IQ3_XS | 2.81GB |
| BioMistral-MedMNX.IQ3_S.gguf | IQ3_S | 2.96GB |
| BioMistral-MedMNX.Q3_K_S.gguf | Q3_K_S | 2.95GB |
| BioMistral-MedMNX.IQ3_M.gguf | IQ3_M | 3.06GB |
| BioMistral-MedMNX.Q3_K.gguf | Q3_K | 3.28GB |
| BioMistral-MedMNX.Q3_K_M.gguf | Q3_K_M | 3.28GB |
| BioMistral-MedMNX.Q3_K_L.gguf | Q3_K_L | 3.56GB |
| BioMistral-MedMNX.IQ4_XS.gguf | IQ4_XS | 3.67GB |
| BioMistral-MedMNX.Q4_0.gguf | Q4_0 | 3.83GB |
| BioMistral-MedMNX.IQ4_NL.gguf | IQ4_NL | 3.87GB |
| BioMistral-MedMNX.Q4_K_S.gguf | Q4_K_S | 3.86GB |
| BioMistral-MedMNX.Q4_K.gguf | Q4_K | 4.07GB |
| BioMistral-MedMNX.Q4_K_M.gguf | Q4_K_M | 4.07GB |
| BioMistral-MedMNX.Q4_1.gguf | Q4_1 | 4.24GB |
| BioMistral-MedMNX.Q5_0.gguf | Q5_0 | 4.65GB |
| BioMistral-MedMNX.Q5_K_S.gguf | Q5_K_S | 4.65GB |
| BioMistral-MedMNX.Q5_K.gguf | Q5_K | 4.78GB |
| BioMistral-MedMNX.Q5_K_M.gguf | Q5_K_M | 4.78GB |
| BioMistral-MedMNX.Q5_1.gguf | Q5_1 | 5.07GB |
| BioMistral-MedMNX.Q6_K.gguf | Q6_K | 5.53GB |
| BioMistral-MedMNX.Q8_0.gguf | Q8_0 | 7.17GB |
Original model description:
license: cc-by-nc-nd-4.0 base_model:
- johnsnowlabs/JSL-MedMNX-7B
- BioMistral/BioMistral-7B-DARE library_name: transformers tags:
- mergekit
- merge
BioMistral-MedMNX
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 johnsnowlabs/JSL-MedMNX-7B 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: johnsnowlabs/JSL-MedMNX-7B
parameters:
density: 0.53
weight: 0.4
- model: BioMistral/BioMistral-7B-DARE
parameters:
density: 0.53
weight: 0.3
merge_method: dare_ties
tokenizer_source: union
base_model: johnsnowlabs/JSL-MedMNX-7B
parameters:
int8_mask: true
dtype: bfloat16
- Downloads last month
- 8
Hardware compatibility
Log In to add your hardware
2-bit
3-bit
4-bit
5-bit
6-bit
8-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support