About

🔑 model key: hive/moderation-11b-vision-language-model

Our Moderation 11B Vision Language Model is built on Llama 3.2 11B Vision Instruct and enhanced with Hive's proprietary data, expanding the capabilities of our established moderation suite.

Designed to handle complex contexts and edge cases, this model combines state-of-the-art multimodal understanding with specialized moderation expertise, making it ideal for identifying and addressing nuanced content across text and images.

How to Get Started

Authentication is required to use these models. You’ll need an API Key, which can be created in the left sidebar.

Follow these steps to generate your key:

Click ‘API Keys’ in the sidebar.
Click ‘+’ to create a new key scoped to your organization. The same key can be used with any "Playground available" model.

⚠️ Important: Keep your API Key secure. Do not expose it in client-side environments like browsers or mobile apps.

Querying Moderation 11B Vision Language Model

Hive AI offers an OpenAI-compatible Rest API for querying LLMs and multimodal LLMs. Here are the ways to call it:

Using the OpenAI SDK
Directly invoking the REST API

Using this API, the model will successively generate new tokens until either the maximum number of output tokens has been reached or if the model’s end-of-sequence (EOS) token has been generated.

Note: Some fields such as top_k are supported via the REST API, but is not supported by the OpenAI SDK.

Using the OpenAI SDK

from openai import OpenAI

# Configure the client with custom base URL and API key
client = OpenAI(
    base_url="https://api.thehive.ai/api/v3/",  # Hive AI's endpoint
    api_key="<YOUR API KEY>"  # Replace with your API key
)

def get_completion(prompt, model = "hive/moderation-11b-vision-language-model"):
  response = client.chat.completions.create(
    model=model,
    messages=[
      {
        "role": "user",
        "content": [
          {"type": "text", "text": prompt},
          {
            "type": "image_url",
            "image_url": {
              "url": "https://d24edro6ichpbm.thehive.ai/example-images/vlm-example-image.jpeg"
            }
          }
        ]
      }
    ],
    temperature=0.7,
    max_tokens=1000
  )

  # Extract the response content
  return response.choices[0].message.content

get_completion("What's in the image?")

const OpenAI = require('openai');

let openai = new OpenAI({
  apiKey: '<YOUR API KEY>',
  baseURL: 'https://api.thehive.ai/api/v3/'
});

async function main() {
  const completion = await openai.chat.completions.create({
    model: 'hive/moderation-11b-vision-language-model',
    messages: [
      {
        role: 'user',
        content: [
          { type: 'text', text: "What's in this image?" },
          {
            type: 'image_url',
            image_url: {
              url: 'https://dash.readme.com/project/thehiveai/v1.0/docs/llama-32-11b-vision-instruct'
            }
          }
        ]
      }
    ],
  });
  console.log(completion.choices[0]?.message?.content);
}

main();

Directly invoking the REST API

All LLMs and Multimodal LLMs support this basic completions REST API.

curl --location --request POST 'https://api.thehive.ai/api/v3/chat/completions' \
--header 'authorization: Bearer <API_KEY>' \
--header 'Content-Type: application/json' \
--data '{
    "model": "hive/moderation-11b-vision-language-model",
    "max_tokens": 1000,
    "messages": [
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": "Can you describe what is in this photo?"
                },
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://d24edro6ichpbm.thehive.ai/example-images/vlm-example-image.jpeg"
                    }
                }
            ]
        }
    ]
}'

After making a request, you’ll receive a JSON response with the generated images as URLs. Here’s a sample output:

{
    "id": "12345-be90-11ef-8469-67890",
    "object": "chat.completion",
    "model": "hive/moderation-11b-vision-language-model",
    "created": 1734671183762,
    "choices": [
        {
            "index": 0,
            "message": {
                "role": "assistant",
                "content": "The image depicts a soccer ball situated on a lush green field, with a stadium in the background. The soccer ball is positioned in the foreground, featuring a white base with vibrant blue, red, and green accents. It rests on a well-manicured grass pitch, which is neatly trimmed and maintained.\n\nIn the background, a large stadium is visible, characterized by its tiered seating and metal framework. The sky above is a brilliant blue, dotted with a few wispy clouds. The overall atmosphere suggests a daytime setting, likely during a match or practice session. The presence of the soccer ball and the stadium implies that the image was taken at a sports venue, possibly during a professional or amateur game."
            },
            "finish_reason": "finish"
        }
    ],
    "usage": {
        "prompt_tokens": 6420,
        "completion_tokens": 142,
        "total_tokens": 6562
    }
}

Schema

Below are the definitions of possible relevant input and output fields. Some fields have default values that will be assigned if the user does not assign a value themselves.

Input

Field	Type	Definition
messages	array of objects	Required. A structured array containing the conversation history. Each object includes a role and content.
model	string	Required. The name of the model to call.
role	string	The role of the participant in the conversation. Must be system, user, or assistant.
content	string OR array of objects	Your content string. If array, each object must have a `type` and corresponding data, as shown in the examples above.
text	string	Referenced inside content arrays, containing the text message to be sent.
image_url	object	Contains the image URL when type is `image_url`. Includes the field `url`.
max_tokens	int	Limits the number of tokens in the output. Default: 2048. Range: 1 to 2048.
temperature	float	Controls randomness in the output. Lower values make output more deterministic. Default: 0.7. Range: [0 to 1]
top_p	float	Nucleus sampling parameter to limit the probability space of token selection. Default: 0.95. Range: [0 to 1].
top_k	int	Limits token sampling to the top K most probable tokens. Default: 0 [0 to 2147483647].

Output

Field	Type	Definition
id	string	The ID of the submitted task.
model	string	The name of the model used.
created	int	The timestamp (in epoch milliseconds) when the task was created.
choices	array of objects	Contains the model’s responses. Each object includes the index, message, and finish_reason.
usage	object	Contains token usage information for the request and response.

Example `messages` array with multiple roles

When submitting messages to the model, you can assign different roles to your content:

system
user
assistant

"messages": [
    {
        "role": "system",
        "content": [
            {
                "type": "text",
                "text": "You are a helpful assistant that provides detailed descriptions of images and answers user queries concisely."
            }
        ]
    },
    {
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": "Can you describe this image in detail?"
            },
            {
                "type": "image_url",
                "image_url": {
                    "url": "https://d24edro6ichpbm.thehive.ai/example-images/vlm-example-image.jpeg"
                }
            }
        ]
    },
    {
        "role": "assistant",
        "content": [
            {
                "type": "text",
                "text": "The image depicts:"
            }
        ]
    }
]

Common Errors

Each model has a default starting rate limit of 5 requests per second. You may see this error below if you submit higher than the rate limit.

To request a higher rate limit please contact [email protected].

{
  "status_code": 429,
  "message": "Too Many Requests"
}

A positive Organization Credit balance is required to continue using Hive Models. Once you run out of credits requests will fail with the following error.

{
  "status_code":405,
  "message":"Your Organization is currently paused. Please check your account balance, our terms and conditions, or contact [email protected] for more information."
}