Update version to 2.0 and add Groq-AI models

README.md

@@ -84,7 +84,7 @@ pip install -r requirements.txt
 
 *Step 1:* **Obtain the HuggingFace API key.**
 
-*Step 2:* Visit the following URL: *https://huggingface.co/settings/tokens* and get your Access Token.
+*Step 2:* Visit the following URL: [HuggingFace Tokens](https://huggingface.co/settings/tokens) and get your Access Token.
 
 *Step 3:* Save the token in a `.env` file as:</br>
 ```bash
@@ -95,7 +95,7 @@ echo "HUGGINGFACE_API_KEY=Your Access Token" > .env
 
 *Step 1:* **Obtain the Google Palm API key.**
 
-*Step 2:* Visit the following URL: *https://makersuite.google.com/app/apikey*
+*Step 2:* Visit the following URL: [Google AI Studio](https://makersuite.google.com/app/apikey)
 
 *Step 3:* Click on the **Create API Key** button.
 
@@ -109,7 +109,7 @@ echo "GEMINI_API_KEY=Your API Key" > .env
 
 *Step 1:* **Obtain the OpenAI API key.**
 
-*Step 2:* Visit the following URL: *https://platform.openai.com/account/api-keys*
+*Step 2:* Visit the following URL: [OpenAI Dashboard](https://platform.openai.com/account/api-keys)
 
 *Step 3:* Sign up for an account or log in if you already have one.
 
@@ -122,6 +122,23 @@ echo "GEMINI_API_KEY=Your API Key" > .env
 echo "OPENAI_API_KEY=Your API Key" > .env
 ```
 
+## Groq AI API Key setup.
+
+*Step 1:* **Obtain the GroqAI API key.**
+
+*Step 2:* Visit the following URL: [Groq AI Console](https://console.groq.com/keys)
+
+*Step 3:* Sign up for an account or log in if you already have one.
+
+*Step 4:* Navigate to the API section in your account.
+
+*Step 5:* Click on the **Create API Key** button.
+
+*Step 6:* The generated key is your API key. Please make sure to **copy** it and **paste** it in the required field below.
+```bash
+echo "GROQ_API_KEY=Your API Key" > .env
+```
+
 # Offline models setup.</br>
 This Interpreter supports offline models via **LM Studio** so to download it from [here](https://lmstudio.ai/) and follow the steps below.
 - Download any model from **LM Studio** like _Phi-2,Code-Llama,Mistral_.
@@ -185,6 +202,8 @@ To use Open-Code-Interpreter, use the following command options:
     - `gpt-4` - Generates code using the GPT 4 model.
     - `gemini-pro` - Generates code using the Gemini Pro model.
     - `palm-2` - Generates code using the PALM 2 model.
+    - `groq-mixtral` - Generates code using the Groq Mixtral model.
+    - `groq-llama2` - Generates code using the Groq Llama2 model.
     - `code-llama` - Generates code using the Code-llama model.
     - `code-llama-phind` - Generates code using the Code-llama Phind model.
     - `mistral-7b` - Generates code using the Mistral 7b model.
@@ -320,6 +339,7 @@ If you're interested in contributing to **Open-Code-Interpreter**, we'd love to
 - **v1.9.1** - Fixed **Unit Tests** and **History Args** <br>
 - **v1.9.2** - Updated **Google Vision** to adapt LiteLLM instead of **Google GenAI** *.<br>
 - **v1.9.3** - Added **Local Models** Support via **LM Studio**.<br>
+- **v2.0** - Added **Groq-AI** Models _Fastest LLM_ with **500 Tokens/Sec** with _Code-LLaMa,Mixtral_ models.<br>
 
 ## 📜 **License**
 

configs/groq-llama2.config

@@ -0,0 +1,17 @@
+# The temperature parameter controls the randomness of the model's output. Lower values make the output more deterministic.
+temperature = 0.1
+
+# The maximum number of new tokens that the model can generate.
+max_tokens = 1024
+
+# The start separator for the generated code.
+start_sep = ```
+
+# The end separator for the generated code.
+end_sep = ```
+
+# If True, the first line of the generated text will be skipped.
+skip_first_line = True
+
+# The model used for generating the code.
+HF_MODEL = groq-llama2

configs/groq-mixtral.config

@@ -0,0 +1,17 @@
+# The temperature parameter controls the randomness of the model's output. Lower values make the output more deterministic.
+temperature = 0.1
+
+# The maximum number of new tokens that the model can generate.
+max_tokens = 1024
+
+# The start separator for the generated code.
+start_sep = ```
+
+# The end separator for the generated code.
+end_sep = ```
+
+# If True, the first line of the generated text will be skipped.
+skip_first_line = True
+
+# The model used for generating the code.
+HF_MODEL = groq-mixtral

interpreter

@@ -29,7 +29,7 @@ def main():
         parser.add_argument('--save_code', '-s', action='store_true', default=False, help='Save the generated code')
         parser.add_argument('--mode', '-md', choices=['code', 'script', 'command','vision','chat'], help='Select the mode (`code` for generating code, `script` for generating shell scripts, `command` for generating single line commands) `vision` for generating text from images')
         parser.add_argument('--model', '-m', type=str, default='code-llama', help='Set the model for code generation. (Defaults to gpt-3.5-turbo)')
-        parser.add_argument('--version', '-v', action='version', version='%(prog)s 1.9.3')
+        parser.add_argument('--version', '-v', action='version', version='%(prog)s 2.0')
         parser.add_argument('--lang', '-l', type=str, default='python', help='Set the interpreter language. (Defaults to Python)')
         parser.add_argument('--display_code', '-dc', action='store_true', default=False, help='Display the code in output')
         parser.add_argument('--history', '-hi', action='store_true', default=False, help='Use history as memory')

interpreter.py

@@ -28,7 +28,7 @@ def main():
         parser.add_argument('--save_code', '-s', action='store_true', default=False, help='Save the generated code')
         parser.add_argument('--mode', '-md', choices=['code', 'script', 'command','vision','chat'], help='Select the mode (`code` for generating code, `script` for generating shell scripts, `command` for generating single line commands) `vision` for generating text from images')
         parser.add_argument('--model', '-m', type=str, default='code-llama', help='Set the model for code generation. (Defaults to gpt-3.5-turbo)')
-        parser.add_argument('--version', '-v', action='version', version='%(prog)s 1.9.3')
+        parser.add_argument('--version', '-v', action='version', version='%(prog)s 2.0')
         parser.add_argument('--lang', '-l', type=str, default='python', help='Set the interpreter language. (Defaults to Python)')
         parser.add_argument('--display_code', '-dc', action='store_true', default=False, help='Display the code in output')
         parser.add_argument('--history', '-hi', action='store_true', default=False, help='Use history as memory')

libs/interpreter_lib.py

@@ -28,7 +28,7 @@
 class Interpreter:
     logger = None
     client = None
-    interpreter_version = "1.9.3"
+    interpreter_version = "2.0"
 
     def __init__(self, args):
         self.args = args
@@ -108,6 +108,7 @@ def initialize_client(self):
 
         self.logger.info(f"Using model {hf_model_name}")
 
+        # checking if the model is from OpenAI
         if "gpt" in self.INTERPRETER_MODEL:
             if os.getenv("OPENAI_API_KEY") is None:
                 load_dotenv()
@@ -121,7 +122,23 @@ def initialize_client(self):
                 raise Exception("OpenAI Key not found in .env file.")
             elif not hf_key.startswith('sk-'):
                 raise Exception("OpenAI token should start with 'sk-'. Please check your .env file.")
+                # checking if the model is from Groq.
 
+        elif "groq" in self.INTERPRETER_MODEL:
+            if os.getenv("GROQ_API_KEY") is None:
+                load_dotenv()
+            if os.getenv("GROQ_API_KEY") is None:
+                # if there is no .env file, try to load from the current working directory
+                load_dotenv(dotenv_path=os.path.join(os.getcwd(), ".env"))
+
+            # Read the token from the .env file
+            groq_key = os.getenv('GROQ_API_KEY')
+            if not groq_key:
+                raise Exception("GroqAI Key not found in .env file.")
+            elif not groq_key.startswith('gsk'):
+                raise Exception("GroqAI token should start with 'gsk'. Please check your .env file.")
+
+        # checking if the model is from Google AI.
         model_api_keys = {
             "palm": "PALM_API_KEY",
             "gemini-pro": "GEMINI_API_KEY"
@@ -143,6 +160,7 @@ def initialize_client(self):
                     raise Exception(f"{api_key_name} not found in .env file.")
                 elif " " in api_key or len(api_key) <= 15:
                     raise Exception(f"{api_key_name} should have no spaces, length greater than 15. Please check your .env file.")
+
         else:
             if os.getenv("HUGGINGFACE_API_KEY") is None:
                 load_dotenv()
@@ -283,7 +301,20 @@ def generate_content(self,message, chat_history: list[tuple[str, str]], temperat
                 self.INTERPRETER_MODEL = "gemini/gemini-pro"
                 response = litellm.completion(self.INTERPRETER_MODEL, messages=messages,temperature=temperature)
                 self.logger.info("Response received from completion function.")
+
+        # Check if the model is Groq-AI
+        elif 'groq' in self.INTERPRETER_MODEL:
 
+            if 'groq-llama2' in self.INTERPRETER_MODEL:
+                self.logger.info("Model is Groq/Llama2.")
+                self.INTERPRETER_MODEL = "groq/llama2-70b-4096"
+            elif 'groq-mixtral' in self.INTERPRETER_MODEL:
+                self.logger.info("Model is Groq/Mixtral.")
+                self.INTERPRETER_MODEL = "groq/mixtral-8x7b-32768"
+
+            response = litellm.completion(self.INTERPRETER_MODEL, messages=messages,temperature=temperature,max_tokens=max_tokens)
+            self.logger.info("Response received from completion function.")
+
         # Check if the model is Local Model
         elif 'local' in self.INTERPRETER_MODEL:
             self.logger.info("Model is Local model")