feat: Adding portkey.ai gateway as a custom model

feat: Adding portkey.ai gateway as a custom model

Author: Fede Kamelhar

docs/user-guide/concepts/model-providers/portkey.md

@@ -0,0 +1,64 @@
+# Portkey 
+
+[Portkey](https://docs.portkey.ai) is a gateway platform that enables seamless integration with multiple language model providers such as OpenAI, Anthropic, and Amazon Bedrock. The Strands Agents SDK includes a Portkey-based provider, allowing you to run agents across OpenAI-compatible models (i.e., those using OpenAI’s API schema) and other supported models via a unified interface.
+
+## Installation
+
+Portkey is configured as an optional dependency in Strands Agents. To install, run:
+
+```bash
+pip install 'strands-agents[portkey]'
+```
+
+## Usage
+
+After installing `portkey`, you can import and initialize the Strands Agents' Portkey provider as follows:
+
+```python
+from strands import Agent
+from strands.models.portkey import PortkeyModel
+from strands_tools import calculator
+
+# Portkey for all models
+model = PortkeyModel(
+    api_key="<PORTKEY_API_KEY>",
+    model_id="anthropic.claude-3-5-sonnet-20241022-v2:0", #Example
+    virtual_key="<BEDROCK_VIRTUAL_KEY>",  # Required for providers like Bedrock. See https://docs.portkey.ai for other provider-specific notes.
+    provider="bedrock",  # You can set the provider to 'anthropic', 'bedrock', 'openai' depending on your model and API setup.
+    base_url="http://portkey-service-gateway.service.prod.example.com/v1",
+)
+
+agent = Agent(model=model, tools=[calculator])
+# You can also add tools like web_search, code_interpreter, etc.
+response = agent("What is 2+2")
+print(response)
+```
+
+
+## Configuration
+
+### Client Configuration
+
+The `client_args` configure the underlying OpenAI client. For a complete list of available arguments, please refer to the OpenAI [source](https://github.com/openai/openai-python).
+
+### Model Configuration
+
+The `model_config` configures the underlying model selected for inference. The supported configurations are:
+
+| Parameter         | Description                          | Example                                    | Options                                                                 |
+|-------------------|--------------------------------------|--------------------------------------------|------------------------------------------------------------------------|
+| `model_id`        | ID of a model to use                | `anthropic.claude-3-5-sonnet-20241022-v2:0` | [reference](https://docs.portkey.ai/docs/llm-routing/overview)              |
+| `base_url`        | Base URL for Portkey service         | `http://portkey-service-gateway.service.prod.example.com/v1` | [reference](https://portkey-ai.com/docs)                               |
+| `provider`        | Model provider                      | `bedrock`                                  | `openai`, `bedrock`, `anthropic`, etc.                                 |
+| `virtual_key`     | Virtual key for authentication       | `<BEDROCK_VIRTUAL_KEY>`                   | [reference](https://portkey-ai.com/docs/authentication)                |
+
+## Troubleshooting
+
+### Module Not Found
+
+If you encounter the error `ModuleNotFoundError: No module named 'portkey'`, this means you haven't installed the `portkey` dependency in your environment. To fix, run `pip install 'strands-agents[portkey]'`.
+
+## References
+
+- [API](../../../api-reference/models.md)
+- [PortkeyAI](https://docs.portkey.ai)

mkdocs.yml

@@ -87,6 +87,7 @@ nav:
         - LlamaAPI: user-guide/concepts/model-providers/llamaapi.md
         - Ollama: user-guide/concepts/model-providers/ollama.md
         - OpenAI: user-guide/concepts/model-providers/openai.md
+        - PortKey: user-guide/concepts/model-providers/portkey.md
         - Custom Providers: user-guide/concepts/model-providers/custom_model_provider.md
       - Streaming:
         - Async Iterators: user-guide/concepts/streaming/async-iterators.md