Add zipnn to readme

README.md

@@ -1,15 +1,58 @@
 ---
 library_name: transformers
 pipeline_tag: image-text-to-text
+license: apache-2.0
+base_model:
+- mistral-community/pixtral-12b
 ---
 
+# Disclaimer and Requirements
+
+This model is a clone of [**mistral-community/pixtral-12b**](https://huggingface.co/mistral-community/pixtral-12b) compressed using ZipNN. Compressed losslessly to 67% its original size, ZipNN saved ~9GB in storage and potentially ~113TB in data transfer **monthly**.
+
+### Requirement
+
+In order to use the model, ZipNN is necessary:
+```bash
+pip install zipnn
+```
+### Use This Model
+```python
+# Load model directly
+from transformers import AutoProcessor, AutoModelForPreTraining
+from zipnn import zipnn_hf
+
+zipnn_hf()
+
+processor = AutoProcessor.from_pretrained("royleibov/pixtral-12b-ZipNN-Compressed")
+model = AutoModelForPreTraining.from_pretrained("royleibov/pixtral-12b-ZipNN-Compressed")
+```
+### ZipNN
+ZipNN also allows you to seemlessly save local disk space in your cache after the model is downloaded.
+
+To compress the cached model, simply run:
+```bash
+python zipnn_compress_path.py safetensors --model royleibov/pixtral-12b-ZipNN-Compressed --hf_cache
+```
+
+The model will be decompressed automatically and safely as long as `zipnn_hf()` is added at the top of the file like in the [example above](#use-this-model).
+
+To decompress manualy, simply run:
+```bash
+python zipnn_decompress_path.py --model royleibov/pixtral-12b-ZipNN-Compressed --hf_cache
+```
+
 # Model Card for Model ID
 Transformers compatible pixtral checkpoints. Make sure to install from source or wait for v4.45! 
 
 ```python
 from PIL import Image
 from transformers import AutoProcessor, LlavaForConditionalGeneration
-model_id = "mistral-community/pixtral-12b"
+from zipnn import zipnn_hf
+
+zipnn_hf()
+
+model_id = "royleibov/pixtral-12b-ZipNN-Compressed"
 model = LlavaForConditionalGeneration.from_pretrained(model_id)
 processor = AutoProcessor.from_pretrained(model_id)
 
@@ -61,7 +104,11 @@ Here's an example with text and multiple images interleaved in the same message:
 ```python
 from PIL import Image
 from transformers import AutoProcessor, LlavaForConditionalGeneration
-model_id = "mistral-community/pixtral-12b"
+from zipnn import zipnn_hf
+
+zipnn_hf()
+
+model_id = "royleibov/pixtral-12b-ZipNN-Compressed"
 model = LlavaForConditionalGeneration.from_pretrained(model_id)
 processor = AutoProcessor.from_pretrained(model_id)
 
@@ -107,5 +154,4 @@ Would you like more information on any specific aspect?
 ```
 
 While it may appear that spacing in the input is disrupted, this is caused by us skipping special tokens for display, and actually "Can this animal" and "live here" are
-correctly separated by image tokens. Try decoding with special tokens included to see exactly what the model sees!
-
+correctly separated by image tokens. Try decoding with special tokens included to see exactly what the model sees!