Update README.md

README.md

@@ -42,7 +42,7 @@ A RoPE base of 500k flattens the rotary frequency spectrum, which gives MotifA1
 ## Installation
 
 ```bash
-pip install codon-model==0.0.6a1
+pip install codon-model==0.0.6a2
 ```
 
 Required artifacts:
@@ -56,6 +56,8 @@ Required artifacts:
 
 ### Load the model
 
+If you have downloaded the `motifa1_sft.safetensors`, then you can load as:
+
 ```python
 from codon.motif import MotifA1
 
@@ -63,16 +65,35 @@ model = MotifA1().load_pretrained('motifa1_sft.safetensors').to('cuda')
 print(model.count_params(human_readable=True))   # -> 105.41 M
 ```
 
+More simply load from network:
+
+```python
+from codon.motif import MotifA1
+
+model = MotifA1().from_remote().to('cuda')
+print(model.count_params(human_readable=True))   # -> 105.41 M
+```
+
 CPU users: replace `'cuda'` with `'cpu'`. Inference works out of the box, just slower.
 
 ### Load the tokenizer
 
+If you have downloaded the `motif.vocab`, then you can load as:
+
 ```python
 from codon.utils.tokens import PackedTokenizer
 
 tokenizer = PackedTokenizer('motif.vocab')
 ```
 
+More simply load from network:
+
+```python
+from codon.motif import MotifA1Tokenizer
+
+tokenizer = MotifA1Tokenizer().from_remote()
+```
+
 ### Streaming chat
 
 ```python