Update README.md

README.md

@@ -8,9 +8,16 @@ pipeline_tag: image-text-to-text
 ---
 
 # Aria-sequential_mlp-bnb_nf4
-BitsAndBytes NF4 quantization from [Aria-sequential_mlp](https://huggingface.co/rhymes-ai/Aria-sequential_mlp), requires about 13.8 GB of VRAM and works on a RTX 3090.
+BitsAndBytes NF4 quantization from [Aria-sequential_mlp](https://huggingface.co/rhymes-ai/Aria-sequential_mlp), requires about 13.8 GB of VRAM and runs on a RTX 3090.
 Currently the model is not 5 GB sharded, as this seems to cause [problems](https://stackoverflow.com/questions/79068298/valueerror-supplied-state-dict-for-layers-does-not-contain-bitsandbytes-an) when loading serialized BNB models. This might make it impossible to load the model in free-tier Colab.
 
+### Installation
+```
+pip install transformers==4.45.0 accelerate==0.34.1 sentencepiece==0.2.0 torchvision requests torch Pillow bitsandbytes
+pip install flash-attn --no-build-isolation
+```
+
+### Inference
 Run this model with:
 ``` python
 import requests
@@ -59,6 +66,7 @@ print(result)
 print(f'Max allocated memory: {torch.cuda.max_memory_allocated(device="cuda") / 1024 ** 3:.3f}GiB')
 ```
 
+### Quantization
 Quantization created with:
 ``` python
 from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig