Add Azure passthrough documentation and sidebar entry (#14958)

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
This commit is contained in:
Teddy Amkie 2025-09-27 10:13:16 -07:00 committed by GitHub
parent a2e72fbcc9
commit fdfff39222
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 90 additions and 0 deletions

View file

@ -0,0 +1,89 @@
# Azure Passthrough
Pass-through endpoints for `/azure`
## Overview
| Feature | Supported | Notes |
|-------|-------|-------|
| Cost Tracking | ❌ | Not supported |
| Logging | ✅ | Works across all integrations |
| Streaming | ✅ | Fully supported |
### When to use this?
- For most use cases, you should use the [native LiteLLM Azure OpenAI Integration](../providers/azure/azure) (`/chat/completions`, `/embeddings`, `/completions`, `/images`, etc.)
- Use this passthrough to call newer or less common Azure OpenAI endpoints that LiteLLM doesn't fully support yet, such as `/assistants`, `/threads`, `/vector_stores`
Simply replace your Azure endpoint (e.g. `https://<your-resource-name>.openai.azure.com`) with `LITELLM_PROXY_BASE_URL/azure`
## Usage Examples
### Assistants API
#### Create Azure OpenAI Client
Make sure you do the following:
- Point `azure_endpoint` to your `LITELLM_PROXY_BASE_URL/azure`
- Use your `LITELLM_API_KEY` as the `api_key`
```python
import openai
client = openai.AzureOpenAI(
azure_endpoint="http://0.0.0.0:4000/azure", # <your-proxy-url>/azure
api_key="sk-anything", # <your-proxy-api-key>
api_version="2024-05-01-preview" # required Azure API version
)
```
#### Create an Assistant
```python
assistant = client.beta.assistants.create(
name="Math Tutor",
instructions="You are a math tutor. Help solve equations.",
model="gpt-4o",
)
```
#### Create a Thread
```python
thread = client.beta.threads.create()
```
#### Add a Message to the Thread
```python
message = client.beta.threads.messages.create(
thread_id=thread.id,
role="user",
content="Solve 3x + 11 = 14",
)
```
#### Run the Assistant
```python
run = client.beta.threads.runs.create(
thread_id=thread.id,
assistant_id=assistant.id,
)
# Check run status
run_status = client.beta.threads.runs.retrieve(
thread_id=thread.id,
run_id=run.id
)
```
#### Retrieve Messages
```python
messages = client.beta.threads.messages.list(
thread_id=thread.id
)
```
#### Delete the Assistant
```python
client.beta.assistants.delete(assistant.id)
```

View file

@ -342,6 +342,7 @@ const sidebars = {
"pass_through/anthropic_completion",
"pass_through/assembly_ai",
"pass_through/bedrock",
"pass_through/azure_passthrough",
"pass_through/cohere",
"pass_through/google_ai_studio",
"pass_through/langfuse",