feat(vertex_ai_anthropic.py): Add support for claude 3 on vertex ai

litellm/llms/custom_httpx/http_handler.py

@@ -0,0 +1,41 @@
+import httpx, asyncio
+from typing import Optional
+
+
+class AsyncHTTPHandler:
+    def __init__(self, concurrent_limit=1000):
+        # Create a client with a connection pool
+        self.client = httpx.AsyncClient(
+            limits=httpx.Limits(
+                max_connections=concurrent_limit,
+                max_keepalive_connections=concurrent_limit,
+            )
+        )
+
+    async def close(self):
+        # Close the client when you're done with it
+        await self.client.aclose()
+
+    async def get(
+        self, url: str, params: Optional[dict] = None, headers: Optional[dict] = None
+    ):
+        response = await self.client.get(url, params=params, headers=headers)
+        return response
+
+    async def post(
+        self,
+        url: str,
+        data: Optional[dict] = None,
+        params: Optional[dict] = None,
+        headers: Optional[dict] = None,
+    ):
+        response = await self.client.post(
+            url, data=data, params=params, headers=headers
+        )
+        return response
+
+    def __del__(self) -> None:
+        try:
+            asyncio.get_running_loop().create_task(self.close())
+        except Exception:
+            pass

litellm/llms/vertex_ai_anthropic.py

@@ -0,0 +1,90 @@
+# What is this?
+## Handler file for calling claude-3 on vertex ai
+from typing import Callable, Optional, Any, Union, List
+import litellm
+
+
+class VertexAIAnthropicConfig:
+    """
+    Reference: https://docs.anthropic.com/claude/reference/messages_post
+
+    Note that the API for Claude on Vertex differs from the Anthropic API documentation in the following ways:
+
+    - `model` is not a valid parameter. The model is instead specified in the Google Cloud endpoint URL.
+    - `anthropic_version` is a required parameter and must be set to "vertex-2023-10-16".
+
+    The class `VertexAIAnthropicConfig` provides configuration for the VertexAI's Anthropic API interface. Below are the parameters:
+
+    - `max_tokens` Required (integer) max tokens,
+    - `anthropic_version` Required (string) version of anthropic for bedrock - e.g. "bedrock-2023-05-31"
+    - `system` Optional (string) the system prompt, conversion from openai format to this is handled in factory.py
+    - `temperature` Optional (float) The amount of randomness injected into the response
+    - `top_p` Optional (float) Use nucleus sampling.
+    - `top_k` Optional (int) Only sample from the top K options for each subsequent token
+    - `stop_sequences` Optional (List[str]) Custom text sequences that cause the model to stop generating
+
+    Note: Please make sure to modify the default parameters as required for your use case.
+    """
+
+    max_tokens: Optional[int] = litellm.max_tokens
+    anthropic_version: Optional[str] = "bedrock-2023-05-31"
+    system: Optional[str] = None
+    temperature: Optional[float] = None
+    top_p: Optional[float] = None
+    top_k: Optional[int] = None
+    stop_sequences: Optional[List[str]] = None
+
+    def __init__(
+        self,
+        max_tokens: Optional[int] = None,
+        anthropic_version: Optional[str] = None,
+    ) -> None:
+        locals_ = locals()
+        for key, value in locals_.items():
+            if key != "self" and value is not None:
+                setattr(self.__class__, key, value)
+
+    @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):
+        return [
+            "max_tokens",
+            "tools",
+            "tool_choice",
+            "stream",
+            "stop",
+            "temperature",
+            "top_p",
+        ]
+
+    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_tokens"] = value
+            if param == "tools":
+                optional_params["tools"] = value
+            if param == "stream":
+                optional_params["stream"] = value
+            if param == "stop":
+                optional_params["stop_sequences"] = value
+            if param == "temperature":
+                optional_params["temperature"] = value
+            if param == "top_p":
+                optional_params["top_p"] = value
+        return optional_params

litellm/model_prices_and_context_window_backup.json

@@ -904,6 +904,22 @@
         "litellm_provider": "vertex_ai-vision-models",
         "mode": "chat"
     },
+    "vertex_ai/claude-3-sonnet@20240229": {
+        "max_tokens": 200000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000003,
+        "output_cost_per_token": 0.000015,
+        "litellm_provider": "vertex_ai",
+        "mode": "chat"
+    },
+    "vertex_ai/claude-3-haiku@20240307": {
+        "max_tokens": 200000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.00000025,
+        "output_cost_per_token": 0.00000125,
+        "litellm_provider": "vertex_ai",
+        "mode": "chat"
+    },
     "textembedding-gecko": {
         "max_tokens": 3072,
         "max_input_tokens": 3072,

model_prices_and_context_window.json

@@ -904,6 +904,22 @@
         "litellm_provider": "vertex_ai-vision-models",
         "mode": "chat"
     },
+    "vertex_ai/claude-3-sonnet@20240229": {
+        "max_tokens": 200000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000003,
+        "output_cost_per_token": 0.000015,
+        "litellm_provider": "vertex_ai",
+        "mode": "chat"
+    },
+    "vertex_ai/claude-3-haiku@20240307": {
+        "max_tokens": 200000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.00000025,
+        "output_cost_per_token": 0.00000125,
+        "litellm_provider": "vertex_ai",
+        "mode": "chat"
+    },
     "textembedding-gecko": {
         "max_tokens": 3072,
         "max_input_tokens": 3072,