OwlChat OpenAi API

Open WebUI provides an OpenAI-compatible Rest API. While working with researchers and students, it was discovered that OpenAI's documentation lacks any description of the REST parameters. If you find that these are incorrect please notify the Research Computing Team via a ticket.
 

OpenWeb UI API:

Open WebUI safeguards its API through features like mandatory authentication using API keys, endpoint restrictions, backend routing for Ollama API requests, granular user permissions, and the ability to configure environment variables to further restrict access, ensuring that only authorized users can interact with the AI models through the API. 

Swagger Documentation:

Users are recommended to test out the API using the provided Swagger docs. https://chat.hpc.fau.edu/docs 

Creating an API Key:

1. Login to https://chat.hpc.fau.edu/  
2. Click on the profile icon on the bottom left or top right
3. Click "Account"
4. Click Show next to the API Keys header and click create API key.

Example Calls: 

Example Completion / Chat:

List Models:

Request:

#!/bin/bash

OPENAI_API_KEY=YOURAPITOKENFROMSETTINGS
curl -X 'GET' \
  'https://chat.hpc.fau.edu/openai/models' \
  -H 'accept: application/json' \
  -H "Authorization: Bearer $OPENAI_API_KEY" | jq

Response:

{
  "data": [
    {
      "id": "llava",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "llava",
      "openai": {
        "id": "llava",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "Llama-3.2-11B-Vision-Instruct",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "Llama-3.2-11B-Vision-Instruct",
      "openai": {
        "id": "Llama-3.2-11B-Vision-Instruct",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "gemma2",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "gemma2",
      "openai": {
        "id": "gemma2",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "codestral",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "codestral",
      "openai": {
        "id": "codestral",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "Llama-3.2-3B-Instruct",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "Llama-3.2-3B-Instruct",
      "openai": {
        "id": "Llama-3.2-3B-Instruct",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "mistral-large",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "mistral-large",
      "openai": {
        "id": "mistral-large",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "tinyllama",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "tinyllama",
      "openai": {
        "id": "tinyllama",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "qwen2",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "qwen2",
      "openai": {
        "id": "qwen2",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    },
    {
      "id": "microsoft/phi-4",
      "object": "model",
      "created": 1677610602,
      "owned_by": "openai",
      "name": "microsoft/phi-4",
      "openai": {
        "id": "microsoft/phi-4",
        "object": "model",
        "created": 1677610602,
        "owned_by": "openai"
      },
      "urlIdx": 1
    }
  ]
}

Chat Completions:

Request:

#!/bin/bash

OPENAI_API_KEY=YOURAPITOKENFROMSETTINGS

curl https://chat.hpc.fau.edu/openai/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -d '{
    "model": "microsoft/phi-4",
    "messages": [
      {
        "role": "developer",
        "content": "You are a helpful assistant."
      },
      {
        "role": "user",
        "content": "Hello!"
      }
    ]
  }' | jq

Response:

{
  "id": "chatcmpl-a3328b366e3848699e0a353e23080987",
  "created": 1740508508,
  "model": "microsoft/phi-4",
  "object": "chat.completion",
  "system_fingerprint": null,
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "content": "Hello! How can I assist you today? Let me know if there's anything specific you'd like help with. 😊",
        "role": "assistant",
        "tool_calls": null,
        "function_call": null,
        "refusal": null,
        "reasoning_content": null
      }
    }
  ],
  "usage": {
    "completion_tokens": 26,
    "prompt_tokens": 19,
    "total_tokens": 45,
    "completion_tokens_details": null,
    "prompt_tokens_details": null
  },
  "service_tier": null,
  "prompt_logprobs": null
}

Example Text-To-Voice using OwlChat.

#!/bin/bash

OPENAI_API_KEY=YOURAPITOKENFROMSETTINGS

OPENAPI_BASE_URL=https://chat.hpc.fau.edu/api

curl $OPENAPI_BASE_URL/v1/audio/speech \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tts-1",
    "input": "Today is a wonderful day to build something people love!",
    "voice": "alloy",
    "response_format": "mp3",
    "speed": 1.0
  }' \
  --output speech.mp3

Prompt Attributes:

Please reference the following curated list of parameters taken from the Azure OpenAI Site. We are validating this list manually to confirm functionality. In the meantime, this is an excellent starting place for users who are interested in manipulating the default settings of different AI models via the REST interface.