add OpenAI o1 config

litellm/llms/OpenAI/o1_reasoning.py

@@ -0,0 +1,72 @@
+"""
+Support for o1 model family 
+
+https://platform.openai.com/docs/guides/reasoning
+
+Translations handled by LiteLLM:
+- modalities: image => drop param (if user opts in to dropping param)  
+- role: system ==> translate to role assistant 
+- streaming => faked by LiteLLM 
+- Tools, response_format =>  drop param (if user opts in to dropping param) 
+- Logprobs => drop param (if user opts in to dropping param) 
+"""
+
+import types
+from typing import Optional, Union
+
+import litellm
+
+from .openai import OpenAIConfig
+
+
+class OpenAIO1Config(OpenAIConfig):
+    """
+    Reference: https://platform.openai.com/docs/guides/reasoning
+
+    """
+
+    @classmethod
+    def get_config(cls):
+        return {
+            k: v
+            for k, v in cls.__dict__.items()
+            if not k.startswith("__")
+            and not isinstance(
+                v,
+                (
+                    types.FunctionType,
+                    types.BuiltinFunctionType,
+                    classmethod,
+                    staticmethod,
+                ),
+            )
+            and v is not None
+        }
+
+    def get_supported_openai_params(self, model: str) -> list:
+        """
+        Get the supported OpenAI params for the given model
+
+        """
+
+        all_openai_params = litellm.OpenAIConfig.get_supported_openai_params(
+            model="gpt-4o"
+        )
+        non_supported_params = [
+            "logprobs",
+            "tools",
+            "tool_choice",
+            "parallel_tool_calls",
+            "function_call",
+            "functions",
+        ]
+
+        return [
+            param for param in all_openai_params if param not in non_supported_params
+        ]
+
+    def map_openai_params(self, non_default_params: dict, optional_params: dict):
+        for param, value in non_default_params.items():
+            if param == "max_tokens":
+                optional_params["max_completion_tokens"] = value
+        return optional_params

litellm/model_prices_and_context_window_backup.json

@@ -11,6 +11,42 @@
         "supports_parallel_function_calling": true,
         "supports_vision": true
     },
+    "o1-preview-2024-09-12": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 32768,
+        "input_cost_per_token": 0.000015,
+        "output_cost_per_token": 0.00006,
+        "litellm_provider": "openai",
+        "mode": "chat"
+    },
+    "o1-preview": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 32768,
+        "input_cost_per_token": 0.000015,
+        "output_cost_per_token": 0.00006,
+        "litellm_provider": "openai",
+        "mode": "chat"
+    },
+    "o1-mini": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 65536,
+        "input_cost_per_token": 0.000003,
+        "output_cost_per_token": 0.000012,
+        "litellm_provider": "openai",
+        "mode": "chat"
+    },
+    "o1-mini-2024-09-12": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 65536,
+        "input_cost_per_token": 0.000003,
+        "output_cost_per_token": 0.000012,
+        "litellm_provider": "openai",
+        "mode": "chat"
+    },
     "gpt-4": {
         "max_tokens": 4096, 
         "max_input_tokens": 8192,
@@ -57,6 +93,54 @@
         "supports_parallel_function_calling": true,
         "supports_vision": true
     },
+    "o1-mini": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000003,
+        "output_cost_per_token": 0.000012,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
+    "o1-mini-2024-09-12": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000003,
+        "output_cost_per_token": 0.000012,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
+    "o1-preview": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000015,
+        "output_cost_per_token": 0.000060,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
+    "o1-preview-2024-09-12": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000015,
+        "output_cost_per_token": 0.000060,
+        "litellm_provider": "openai",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true
+    },
     "chatgpt-4o-latest": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,