Purpose of this document

Currently, we have a very minimal and basic API. We are looking to expand more in the future but wanted to create a quick document highlighting the available endpoints today.

Note

• Plus Tier subscription ($15/month)

API Keys

You can find your API keys on the https://infermatic.ai/ website, after you login. They are in the left navbar below the Join Discord button

In red you have some recommendations for the settings, so the API do a good performance

Conventions

The base URL to send all API requests is https://api.totalgpt.ai/v1. HTTPS is required for all API requests.

The Infermatic API follows RESTful conventions when possible, with most operations performed via GET, POST requests on page and database resources. Request and response bodies are encoded as JSON.

Request Limits

We don’t have a limit for the tokens, however we do have a limitation of requests per minute.

• 18/minute.
• 2 parallel requests.

Size limits UI

Here are the context length for the model, remember that if you have a recommendation about any length (be that you want to increase or decrease) you can request it on our Discord

Status Codes

HTTP response codes are used to indicate general classes of success.

Success Code

Sample success response

GET

{
    "data": [
        {
            "id": "Noromaid-v0.1-mixtral-8x7b-Instruct-v3",
            "object": "model",
            "created": 1677610602,
            "owned_by": "openai"
        },
        {
            "id": "bagel-dpo-8x7b-v0.2",
            "object": "model",
            "created": 1677610602,
            "owned_by": "openai"
        },
        {
            "id": "Mixtral-8x7B-Instruct-v0.1",
            "object": "model",
            "created": 1677610602,
            "owned_by": "openai"
        },
        {
            "id": "Noromaid-13b-v0.2",
            "object": "model",
            "created": 1677610602,
            "owned_by": "openai"
        },
        {
            "id": "MythoMax-L2-13b",
            "object": "model",
            "created": 1677610602,
            "owned_by": "openai"
        }
    ],
    "object": "list"
}

POST

{
    "id": "chatcmpl-f7fdc258-16f1-42bf-bbc0-75bd553bd93d",
    "choices": [
        {
            "finish_reason": "eos_token",
            "index": 0,
            "message": {
                "content": " I am a artificial intelligence language model, designed to understand and respond to natural language inputs. I can provide information, answer questions, and engage in conversation on a wide range of topics. However, I do not have personal experiences or emotions, and I do not exist in the physical world. I am a computer program running on a server, and my capabilities are limited to what has been programmed into me.",
                "role": "assistant"
            }
        }
    ],
    "created": 1705445740,
    "model": "mistralai/Mixtral-8x7B-Instruct-v0.1",
    "object": "chat.completion",
    "system_fingerprint": null,
    "usage": {
        "prompt_tokens": 14,
        "completion_tokens": 81,
        "total_tokens": 95
    }
}

Endpoints

You can test it on Postman

Endpoints

Where to use?

There are plenty of usages for our API thats why we made guides for How to use or integrate the API, we’ve seen this are the most common chatbots used by our community:

Silly Tavern

https://svak.notion.site/Silly-Tavern-GUIDE-06548e9285414b66a5515e5c92faf99f?pvs=4

AMICA

https://svak.notion.site/svak/AMICA-API-GUIDE-42d95835c8cb416eae6b2cdf3287da1f

AGNAI

https://svak.notion.site/svak/AGNAI-API-GUIDE-ce250f04fd4940a48abe327b152069eb

HELP

Sure! We will be very happy to hear all the questions and recommendations you have :)

So, join our discord and lets get on it!!!

And have in mind that:

• The #api channel is for API related questions
• The #questions channel is more general, about our service as whole
• Speak up! In the #feedback channel you can complain, recommend or suggest things about anything. Such as the API, models, UI, service, discord channels, documentation… etc
• We see you! Don’t be shy, you can add your creations on #random-stuff
• If you need help configuring something or consider that your question is going to be long, feel free to create a thread on #ask-for-help

—> https://discord.gg/infermaticai

With all this said, thanks for letting us be part of your journey.

GL and let the fun start!!

Warm regards,

- InferTeam