Update constants/models.py

constants/models.py

@@ -14,7 +14,8 @@ MODEL_MAP = {
     "qwen2-1.5b-instruct" : "Qwen/Qwen2-1.5B-Instruct",
     "llama3-8b" : "meta-llama/Meta-Llama-3-8B-Instruct",
     "mistral-nemo-instruct":"mistralai/Mistral-Nemo-Instruct-2407",
-    "zephyr-7b-beta":"HuggingFaceH4/zephyr-7b-beta"
+    "zephyr-7b-beta":"HuggingFaceH4/zephyr-7b-beta",
+    "phi-3-mini-instruct":"microsoft/Phi-3-mini-4k-instruct"
 }
 
 AVAILABLE_MODELS = list(MODEL_MAP.keys())
@@ -45,7 +46,8 @@ STOP_SEQUENCES_MAP = {
     "llama3-8b" : "<|eot_id|>",
     "llama3-70b" : "<|eot_id|>",
     "mistral-nemo-instruct": "</s>",
-    "zephyr-7b-beta":"</s>"
+    "zephyr-7b-beta":"</s>",
+    "phi-3-mini-instruct":"<|end|>"
 }
 
 TOKEN_LIMIT_MAP = {
@@ -65,6 +67,7 @@ TOKEN_LIMIT_MAP = {
     "qwen2-7b-instruct": 32768,
     "qwen2-1.5b-instruct": 32768,
     "llama3-8b": 8192,
+    "phi-3-mini-instruct":4096
 }
 
 TOKEN_RESERVED = 20
@@ -100,7 +103,7 @@ LLM_MODELS_DICTS = [
         "object": "model",
         "created": 1700000000,
         "owned_by": "01-ai",
-        "type": "pro",
+        "type": "basic",
     },
     {
         "id": "gemma-7b",
@@ -190,6 +193,14 @@ LLM_MODELS_DICTS = [
         "owned_by": "HuggingFaceH4",
         "type": "basic",
     },
+    {
+        "id": "phi-3-mini-instruct",
+        "description": "[microsoft/Phi-3-mini-4k-instruct]: https://huggingface.co/microsoft/Phi-3-mini-4k-instruct",
+        "object": "model",
+        "created": 1700000000,
+        "owned_by": "microsoft",
+        "type": "basic",
+    },
 ]