Add Mixtral

README.md

@@ -9,26 +9,22 @@
 
 [![Streamlit App](https://static.streamlit.io/badges/streamlit_badge_black_white.svg)](https://snowchat.streamlit.app/)
 
-
 ![156shots_so](https://github.com/kaarthik108/snowChat/assets/53030784/7538d25b-a2d4-4a2c-9601-fb4c7db3c0b6)
 
-
 **snowChat** is an intuitive and user-friendly application that allows users to interact with their Snowflake data using natural language queries. Type in your questions or requests, and SnowChat will generate the appropriate SQL query and return the data you need. No more complex SQL queries or digging through tables - SnowChat makes it easy to access your data! By bringing data one step closer, SnowChat empowers users to make data-driven decisions faster and more efficiently, reducing the barriers between users and the insights they seek.
 
 ## Supported LLM's
+
 - GPT-3.5-turbo-16k
-- Code-llama-13B-instruct
 - Claude-instant-v1
+- Mixtral 8x7B
 
 #
 
-
 https://github.com/kaarthik108/snowChat/assets/53030784/24105e23-69d3-4676-b6d6-d8157dd1580a
 
-
 #
 
-
 ## 🌟 Features
 
 - **Conversational AI**: Harnesses ChatGPT to translate natural language into precise SQL queries.
@@ -37,7 +33,6 @@ https://github.com/kaarthik108/snowChat/assets/53030784/24105e23-69d3-4676-b6d6-
 - **Self-healing SQL**: Proactively suggests solutions for SQL errors, streamlining data access.
 - **Interactive User Interface**: Transforms data querying into an engaging conversation, complete with a chat reset option.
 
-
 ## 🛠️ Installation
 
 1. Clone this repository:
@@ -47,7 +42,7 @@ https://github.com/kaarthik108/snowChat/assets/53030784/24105e23-69d3-4676-b6d6-
    cd snowchat
    pip install -r requirements.txt
 
-3. Set up your `OPENAI_API_KEY`, `ACCOUNT`, `USER_NAME`, `PASSWORD`, `ROLE`, `DATABASE`, `SCHEMA`,  `WAREHOUSE`, `SUPABASE_URL` , `SUPABASE_SERVICE_KEY` and `REPLICATE_API_TOKEN` in project directory `secrets.toml`.
+3. Set up your `OPENAI_API_KEY`, `ACCOUNT`, `USER_NAME`, `PASSWORD`, `ROLE`, `DATABASE`, `SCHEMA`, `WAREHOUSE`, `SUPABASE_URL` , `SUPABASE_SERVICE_KEY` and `REPLICATE_API_TOKEN` in project directory `secrets.toml`.
 
 4. Make you're schemas and store them in docs folder that matches you're database.
 
@@ -68,7 +63,6 @@ https://github.com/kaarthik108/snowChat/assets/53030784/24105e23-69d3-4676-b6d6-
 
 [![Star History Chart](https://api.star-history.com/svg?repos=kaarthik108/snowChat&type=Date)]
 
-
 ## 🤝 Contributing
 
 Feel free to contribute to this project by submitting a pull request or opening an issue. Your feedback and suggestions are greatly appreciated!

chain.py

@@ -6,7 +6,7 @@
 from langchain.chains.question_answering import load_qa_chain
 from langchain.chat_models import ChatOpenAI
 from langchain.embeddings.openai import OpenAIEmbeddings
-from langchain.llms import OpenAI, Replicate
+from langchain.llms import OpenAI
 from langchain.llms.bedrock import Bedrock
 from langchain.vectorstores import SupabaseVectorStore
 from pydantic import BaseModel, validator
@@ -18,9 +18,6 @@
 supabase_key = st.secrets["SUPABASE_SERVICE_KEY"]
 supabase: Client = create_client(supabase_url, supabase_key)
 
-VERSION = "1f01a52ff933873dff339d5fb5e1fd6f24f77456836f514fa05e91c1a42699c7"
-LLAMA = "meta/codellama-13b-instruct:{}".format(VERSION)
-
 
 class ModelConfig(BaseModel):
     model_type: str
@@ -29,7 +26,7 @@ class ModelConfig(BaseModel):
 
     @validator("model_type", pre=True, always=True)
     def validate_model_type(cls, v):
-        if v not in ["code-llama", "gpt", "claude"]:
+        if v not in ["gpt", "claude", "mixtral"]:
             raise ValueError(f"Unsupported model type: {v}")
         return v
 
@@ -42,26 +39,12 @@ def __init__(self, config: ModelConfig):
         self.setup()
 
     def setup(self):
-        if self.model_type == "code-llama":
-            self.setup_llama()
-        elif self.model_type == "gpt":
+        if self.model_type == "gpt":
             self.setup_gpt()
         elif self.model_type == "claude":
             self.setup_claude()
-
-    def setup_llama(self):
-        self.q_llm = Replicate(
-            model=LLAMA,
-            input={"temperature": 0.2, "max_length": 200, "top_p": 1},
-            replicate_api_token=self.secrets["REPLICATE_API_TOKEN"],
-        )
-        self.llm = Replicate(
-            streaming=True,
-            callbacks=[self.callback_handler],
-            model=LLAMA,
-            input={"temperature": 0.2, "max_length": 300, "top_p": 1},
-            replicate_api_token=self.secrets["REPLICATE_API_TOKEN"],
-        )
+        elif self.model_type == "mixtral":
+            self.setup_mixtral()
 
     def setup_gpt(self):
         self.q_llm = OpenAI(
@@ -80,6 +63,25 @@ def setup_gpt(self):
             streaming=True,
         )
 
+    def setup_mixtral(self):
+        self.q_llm = OpenAI(
+            temperature=0.1,
+            openai_api_key=self.secrets["MIXTRAL_API_KEY"],
+            model_name="mistralai/Mixtral-8x7B-Instruct-v0.1",
+            max_tokens=500,
+            base_url="https://api.together.xyz/v1",
+        )
+
+        self.llm = ChatOpenAI(
+            model_name="mistralai/Mixtral-8x7B-Instruct-v0.1",
+            temperature=0.5,
+            openai_api_key=self.secrets["MIXTRAL_API_KEY"],
+            max_tokens=500,
+            callbacks=[self.callback_handler],
+            streaming=True,
+            base_url="https://api.together.xyz/v1",
+        )
+
     def setup_claude(self):
         bedrock_runtime = boto3.client(
             service_name="bedrock-runtime",
@@ -132,8 +134,10 @@ def load_chain(model_name="GPT-3.5", callback_handler=None):
         model_type = "claude"
     elif "GPT-3.5" in model_name:
         model_type = "gpt"
+    elif "mixtral" in model_name.lower():
+        model_type = "mixtral"
     else:
-        model_type = "code-llama"
+        raise ValueError(f"Unsupported model name: {model_name}")
 
     config = ModelConfig(
         model_type=model_type, secrets=st.secrets, callback_handler=callback_handler

main.py

@@ -17,7 +17,7 @@
 st.caption("Talk your way through data")
 model = st.radio(
     "",
-    options=["✨ GPT-3.5", "🐐 code-LLama", "♾️ Claude"],
+    options=["✨ GPT-3.5", "♾️ Claude", "⛰️ Mixtral"],
     index=0,
     horizontal=True,
 )

requirements.txt

@@ -1,4 +1,4 @@
-langchain==0.0.305
+langchain==0.0.350
 pandas==1.5.0
 pydantic==1.10.8
 snowflake_snowpark_python==1.5.0