diff --git a/docs/language-models/hosted-models/groq.mdx b/docs/language-models/hosted-models/groq.mdx
new file mode 100644
index 0000000000..f0c151e46b
--- /dev/null
+++ b/docs/language-models/hosted-models/groq.mdx
@@ -0,0 +1,72 @@
+---
+title: Groq
+---
+
+To use Open Interpreter with a model from Groq, simply run:
+
+<CodeGroup>
+
+```bash Terminal
+interpreter --model groq/llama3-8b-8192
+```
+
+```python Python
+from interpreter import interpreter
+
+interpreter.llm.model = "groq/llama3-8b-8192"
+interpreter.llm.api_key = '<groq-api-key>'
+interpreter.chat()
+```
+
+</CodeGroup>
+
+If you are having any issues when passing the `--model`, try adding the `--api_base`:
+
+<CodeGroup>
+
+```bash Terminal
+interpreter --api_base "https://api.groq.com/openai/v1" --model groq/llama3-8b-8192 --api_key $GROQ_API_KEY
+```
+
+```python Python
+from interpreter import interpreter
+
+interpreter.llm.model = "groq/llama3-8b-8192"
+interpreter.llm.api_key = '<groq-api-key>'
+interpreter.llm.api_base = "https://api.groq.com/openai/v1"
+interpreter.llm.context_window = 32000
+interpreter.chat()
+```
+
+</CodeGroup>
+
+# Supported Models
+
+We support any model on [Groq's models page:](https://console.groq.com/docs/models)
+
+<CodeGroup>
+
+```bash Terminal
+interpreter --model groq/mixtral-8x7b-32768
+interpreter --model groq/llama3-8b-8192
+interpreter --model groq/llama3-70b-8192
+interpreter --model groq/gemma-7b-it
+```
+
+```python Python
+interpreter.llm.model = "groq/mixtral-8x7b-32768"
+interpreter.llm.model = "groq/llama3-8b-8192"
+interpreter.llm.model = "groq/llama3-70b-8192"
+interpreter.llm.model = "groq/gemma-7b-it"
+```
+
+</CodeGroup>
+
+# Required Environment Variables
+
+Run `export GROQ_API_KEY='<your-key-here>'` or place it in your rc file and re-source
+Set the following environment variables [(click here to learn how)](https://chat.openai.com/share/1062cdd8-62a1-4aa8-8ec9-eca45645971a) to use these models.
+
+| Environment Variable | Description                                          | Where to Find                                                       |
+| -------------------- | ---------------------------------------------------- | ------------------------------------------------------------------- |
+| `GROQ_API_KEY`     | The API key for authenticating to Groq's services. | [Groq Account Page](https://console.groq.com/keys) |
diff --git a/interpreter/core/llm/llm.py b/interpreter/core/llm/llm.py
index a539e59a8c..35bcdae472 100644
--- a/interpreter/core/llm/llm.py
+++ b/interpreter/core/llm/llm.py
@@ -1,4 +1,8 @@
 import litellm
+from groq import Groq
+
+groq_client = [None]
+
 import tokentrim as tt
 
 from ...terminal_interface.utils.display_markdown_message import (
@@ -9,6 +13,7 @@
 from .utils.convert_to_openai_messages import convert_to_openai_messages
 
 litellm.suppress_debug_info = True
+import os
 import time
 
 
@@ -26,6 +31,7 @@ def __init__(self, interpreter):
 
         # Settings
         self.model = "gpt-4-turbo"
+        self.model = "groq/mixtral-8x7b-32768"  # can now use models from groq. `export GROQ_API_KEY="your-key-here")` or use --model
         self.temperature = 0
         self.supports_vision = False
         self.supports_functions = None  # Will try to auto-detect
@@ -67,7 +73,7 @@ def run(self, messages):
                     self.supports_functions = False
             except:
                 self.supports_functions = False
-            
+
         # Trim image messages if they're there
         if self.supports_vision:
             image_messages = [msg for msg in messages if msg["type"] == "image"]
@@ -210,7 +216,54 @@ def fixed_litellm_completions(**params):
     # Run completion
     first_error = None
     try:
-        yield from litellm.completion(**params)
+        def source(**params):
+            """Get Completions Using LiteLLM"""
+            yield from litellm.completion(**params)
+
+        if "model" in params and "groq/" in params["model"]:
+
+            def groq_complete(**params):
+                if groq_client[0] is None:
+                    groq_client[0] = Groq(
+                        # This is the default and can be omitted
+                        api_key=os.environ.get("GROQ_API_KEY"),
+                        timeout=2,
+                        max_retries=3,
+                    )
+                res = (
+                    groq_client[0]
+                    .chat.completions.create(
+                        messages=params["messages"],
+                        model=params["model"].split("groq/")[1],
+                    )
+                    .choices[0]
+                    .message.content
+                )
+                return res
+
+            def s(**params):
+                """Get Completions Using Groq"""
+                params["stream"] = (
+                    False  # To keep things simple for now, and groq is super fast anyway
+                )
+                word_by_word = False
+                if word_by_word:
+                    for chunk in groq_complete(**params).split(" "):
+                        yield {
+                            "choices": [
+                                {"delta": {"type": "message", "content": chunk + " "}}
+                            ]
+                        }
+                else:
+                    for whole in [groq_complete(**params)]:
+                        yield {
+                            "choices": [
+                                {"delta": {"type": "message", "content": whole}}
+                            ]
+                        }
+
+            source = s
+        yield from source(**params)
     except Exception as e:
         # Store the first error
         first_error = e
diff --git a/pyproject.toml b/pyproject.toml
index 79809ccdc7..f6474da6ea 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -29,6 +29,7 @@ matplotlib = "^3.8.2"
 toml = "^0.10.2"
 posthog = "^3.1.0"
 tiktoken = "^0.6.0"
+groq = "^4.3.0"
 
 
 #Optional dependencies