fix dependencies, improve readme

air_llm/README.md

@@ -111,7 +111,9 @@ When initialize the model, we support the following configurations:
 
 ### 5. Supported Models
 
-#### [HF open llm leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) top models
+#### [HF open llm leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) top models 
+
+**Including but not limited to the following:** (Most of the open models are based on llama2, so should be supported by default)
 
 @12/01/23
 
@@ -131,6 +133,8 @@ When initialize the model, we support the following configurations:
 
 #### [opencompass leaderboard](https://opencompass.org.cn/leaderboard-llm) top models
 
+**Including but not limited to the following:** (Most of the open models are based on llama2, so should be supported by default)
+
 @12/01/23
 
 | Rank  | Model | Supported |

air_llm/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r") as fh:
 
 setuptools.setup(
     name="airllm",
-    version="2.1.0",
+    version="2.1.1",
     author="Gavin Li",
     author_email="gavinli@animaai.cloud",
     description="AirLLM allows single 4GB GPU card to run 70B large language models without quantization, distillation or pruning.",
@@ -20,7 +20,7 @@ setuptools.setup(
         'accelerate',
         'safetensors',
         'optimum',
-        'huggingface-hub'
+        'huggingface-hub',
         'scipy',
         #'bitsandbytes' set it to optional to support fallback when not installable
     ],