add support for friendli dedicated endpoint

docs/my-website/docs/providers/friendliai.md

@@ -0,0 +1,60 @@
+# FriendliAI
+https://suite.friendli.ai/
+
+**We support ALL FriendliAI models, just set `friendliai/` as a prefix when sending completion requests**
+
+## API Key
+```python
+# env variable
+os.environ['FRIENDLI_TOKEN']
+os.environ['FRIENDLI_API_BASE'] # Optional. Set this when using dedicated endpoint.
+```
+
+## Sample Usage
+```python
+from litellm import completion
+import os
+
+os.environ['FRIENDLI_TOKEN'] = ""
+response = completion(
+    model="friendliai/mixtral-8x7b-instruct-v0-1", 
+    messages=[
+       {"role": "user", "content": "hello from litellm"}
+   ],
+)
+print(response)
+```
+
+## Sample Usage - Streaming
+```python
+from litellm import completion
+import os
+
+os.environ['FRIENDLI_TOKEN'] = ""
+response = completion(
+    model="friendliai/mixtral-8x7b-instruct-v0-1", 
+    messages=[
+       {"role": "user", "content": "hello from litellm"}
+   ],
+    stream=True
+)
+
+for chunk in response:
+    print(chunk)
+```
+
+
+## Supported Models
+### Serverless Endpoints
+We support ALL FriendliAI AI models, just set `friendliai/` as a prefix when sending completion requests
+
+| Model Name               | Function Call                                                                                                                                                      |
+|--------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| mixtral-8x7b-instruct | `completion(model="friendliai/mixtral-8x7b-instruct-v0-1", messages)` | 
+| meta-llama-3-8b-instruct | `completion(model="friendliai/meta-llama-3-8b-instruct", messages)` |
+| meta-llama-3-70b-instruct | `completion(model="friendliai/meta-llama-3-70b-instruct", messages)` |  
+
+### Dedicated Endpoints
+```
+model="friendliai/$ENDPOINT_ID:$ADAPTER_ROUTE"
+```

docs/my-website/sidebars.js

@@ -158,6 +158,7 @@ const sidebars = {
         "providers/triton-inference-server",
         "providers/ollama", 
         "providers/perplexity", 
+        "providers/friendliai",
         "providers/groq", 
         "providers/deepseek", 
         "providers/fireworks_ai",

litellm/utils.py

@@ -4486,7 +4486,10 @@ def get_llm_provider(
                     or get_secret("TOGETHER_AI_TOKEN")
                 )
             elif custom_llm_provider == "friendliai":
-                api_base = "https://inference.friendli.ai/v1"
+                api_base = (
+                    get_secret("FRIENDLI_API_BASE")
+                    or "https://inference.friendli.ai/v1"
+                )
                 dynamic_api_key = (
                     api_key
                     or get_secret("FRIENDLIAI_API_KEY")