diff --git a/.env.example b/.env.example
index 21daf4db..7f8bb5d9 100644
--- a/.env.example
+++ b/.env.example
@@ -49,3 +49,6 @@ AGNES_API_KEY = "<YOUR_KEY>"
 
 # SILICONFLOW_API_KEY
 SILICONFLOW_API_KEY = "<YOUR_KEY>"
+
+# Groq API Configuration
+GROQ_API_KEY = "<YOUR_KEY>"
diff --git a/llm_inference/model_inference.py b/llm_inference/model_inference.py
index ad231c18..a2baa35c 100644
--- a/llm_inference/model_inference.py
+++ b/llm_inference/model_inference.py
@@ -33,6 +33,7 @@ def __init__(self):
         self.deepseek_api_key = os.getenv("DEEPSEEK_API_KEY")
         self.perplexity_api_key = os.getenv("PERPLEXITY_API_KEY")
         self.replicate_api_key = os.getenv("REPLICATE_API_KEY")
+        self.groq_api_key = os.getenv("GROQ_API_KEY")
 
         # AWS credentials
         self.aws_access_key_id = os.getenv("AWS_ACCESS_KEY_ID")
@@ -97,7 +98,8 @@ def infer(
                     return self._call_agnes(model_name, prompt)
                 elif provider == "siliconflow":
                     return self._call_siliconflow(model_name, prompt)
-
+                elif provider == "groq":
+                    return self._call_groq(model_name, prompt)
                 else:
                     # Default to Together API for most open-source models
                     return self._call_together(model_name, prompt)
@@ -217,6 +219,12 @@ def _get_provider(self, model_name: str) -> str:
             # Siliconflow
             "THUDM/GLM-4-9B-0414": "siliconflow",
             "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B": "siliconflow",
+            # Groq
+            "llama-3.3-70b-versatile": "groq",
+            "llama-3.1-8b-instant": "groq",
+            "llama-3.1-70b-versatile": "groq",
+            "mixtral-8x7b-32768": "groq",
+            "gemma2-9b-it": "groq",
         }
 
         # Check if exact model name is in mapping
@@ -920,3 +928,42 @@ def _call_aws(self, model_name: str, prompt: str) -> Dict[str, Any]:
                 "model_used": model_name,
                 "provider": "aws",
             }
+
+    def _call_groq(self, model_name: str, prompt: str) -> Dict[str, Any]:
+        """Call Groq API."""
+        import openai
+
+        # Groq uses OpenAI-compatible API
+        client = openai.OpenAI(
+            api_key=self.groq_api_key, base_url="https://api.groq.com/openai/v1"
+        )
+
+        response = client.chat.completions.create(
+            model=model_name,
+            messages=[{"role": "user", "content": prompt}],
+            max_tokens=2048,
+            temperature=0.7,
+        )
+
+        usage = getattr(response, "usage", None)
+        input_tokens = getattr(usage, "prompt_tokens", 0) if usage is not None else 0
+        completion_tokens = (
+            getattr(usage, "completion_tokens", 0) if usage is not None else 0
+        )
+        total_tokens = (
+            getattr(usage, "total_tokens", 0)
+            if usage is not None
+            else input_tokens + completion_tokens
+        )
+
+        return {
+            "response": response.choices[0].message.content,
+            "success": True,
+            "token_usage": {
+                "input_tokens": input_tokens,
+                "output_tokens": completion_tokens,
+                "total_tokens": total_tokens,
+            },
+            "model_used": model_name,
+            "provider": "groq",
+        }
diff --git a/model_cost/model_cost.json b/model_cost/model_cost.json
index ae82c1d1..a1848863 100644
--- a/model_cost/model_cost.json
+++ b/model_cost/model_cost.json
@@ -386,5 +386,25 @@
   "google/gemma-4-31b-it": {
     "input_token_price_per_million": 0.08,
     "output_token_price_per_million": 0.35
+  },
+  "llama-3.3-70b-versatile": {
+    "input_token_price_per_million": 0.59,
+    "output_token_price_per_million": 0.79
+  },
+  "llama-3.1-8b-instant": {
+    "input_token_price_per_million": 0.05,
+    "output_token_price_per_million": 0.08
+  },
+  "llama-3.1-70b-versatile": {
+    "input_token_price_per_million": 0.59,
+    "output_token_price_per_million": 0.79
+  },
+  "mixtral-8x7b-32768": {
+    "input_token_price_per_million": 0.24,
+    "output_token_price_per_million": 0.24
+  },
+  "gemma2-9b-it": {
+    "input_token_price_per_million": 0.20,
+    "output_token_price_per_million": 0.20
   }
 }
diff --git a/universal_model_names.py b/universal_model_names.py
index 884554a2..ae7a6877 100644
--- a/universal_model_names.py
+++ b/universal_model_names.py
@@ -138,6 +138,12 @@
     "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
     # KT-ModelRouter pool additions
     "google/gemma-4-31b-it",
+    # Groq
+    "llama-3.3-70b-versatile",
+    "llama-3.1-8b-instant",
+    "llama-3.1-70b-versatile",
+    "mixtral-8x7b-32768",
+    "gemma2-9b-it",
 ]