Text Generation APIs

Overview

Hive provides simple APIs to LLMs (large language models) to produce text from a given prompt. These models produce writing across many genres and formats for a wide variety of use cases, including question answering, story writing, conversations, and coding.

From prompts, these models can generate text responses in code format, structured JSON data, or human-like diction.

For our initial batch of LLM APIs, we've focused on Llama's open source suite, which allow developers to achieve great quality and performance on text-based applications at a fraction of the cost.

Models

We offer a variety of Meta’s open-source Llama Instruct models from the 3.1 and 3.2 series, with additional models to be served in the near future.

Here are the differences between our current LLM offerings:

Model	Description
Llama 3.2 1B Instruct	Llama 3.2 1B Instruct is a lightweight, multilingual, text-only model that fits onto both edge and mobile devices. Use cases where the model excels include summarizing or rewriting inputs, as well as instruction following. Try out the Playground for this model here !
Llama 3.2 3B Instruct	A versatile, high-performance model optimized for lightweight language processing across diverse applications. Try out the Playground for this model here !
Llama 3.1 8B Instruct	Llama 3.1 8B Instruct is a more advanced multilingual model designed for complex integrations, offering exceptional language understanding and accuracy. Try out the Playground for this model here !
Llama 3.1 70B Instruct	A high-powered multilingual model built for large-scale applications, delivering superior performance for intricate tasks and workflows. Compared to other available open source and closed chat models, Llama 3.1 70B Instruct achieves higher scores across common industry benchmarks. Try out the Playground for this model here !

Getting your V3 API Key

Your V3 API Key can be created in the left sidebar of the Hive UI, under "API Keys."

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.

V3 Request Format

Below are the input fields for a multimodal LLM cURL request. The asterisk (*) next to an input field designates that it is required.

model*: The name of the model to call.
max_tokens: Limits the number of tokens in the output. Default: 2048. Range: 1 to 2048.
messages*: A structured array containing the conversation history. Each object includes a role and content.
role: The role of the participant in the conversation. Must be system, user, or assistant.
content: Your content string, which can be a string or an array of objects. If it's an array, each object must have a type and corresponding data, as shown in the example below.
text: Referenced inside content arrays, containing the text message to be sent.

Here is an example of a cURL request using the following format:

curl --location --request POST 'https://api.thehive.ai/api/v3/chat/completions' \
--header 'authorization: Bearer <API_KEY>' \
--header 'Content-Type: application/json' \
--data '{
    "model": "meta-llama/llama-3.2-1b-instruct",
    "max_tokens": 1000,
    "messages": [
        {
            "role": "user",
            "content": "Explain the concept of reinforcement learning."
        }
    ]
}'

Response

To see example API responses for these models, you can visit their respective API playgrounds, which contain additional documentation.

LEGACY -- V2 Request Format

Below are the input fields for an LLM cURL request. The asterisk (*) next to an input field designates that it is required.

Input Field	Type	Definition
text_data*	string	The main text prompt. This describes what the text response should include.
callback_url*	string	When the task is completed, we will send a callback from our servers to this callback url.
system_prompt	string	String that provides context for how the model should respond to all requests.
prompt_history		Provides the chat history in chronological order, where the last item is the most recent chat. Each item needs to be a dictionary with the keys "content" and "role".
top_p	float	A value between 0 and 1. If the value is 0, requests should be more diverse than when set to a higher value.
temperature	float	A value between 0 and 1. If this value is 0, a repeated request should yield the same response. If this value is 1, a repeated request should yield different responses.
max_tokens	int	This is the maximum token window that you want for your model completion output.

Here is an example of a cURL request using the following format:

curl --location 'https://api.thehive.ai/api/v2/task/async' \
--header 'Authorization: Token <YOUR_API_KEY>' \
--header 'Content-Type: application/json' \
--data '{
    "text_data": "<YOUR_PROMPT>",
    "callback_url": "<YOUR_CALLBACK_URL>",
    "options": {
        "system_prompt": "<YOUR_SYSTEM_PROMPT>",
        "prompt_history": "<YOUR_PROMPT_HISTORY>",
        "top_p": <YOUR_TOP_P>,
        "temperature": <YOUR_TEMPERATURE>,
        "max_tokens": <YOUR_MAX_TOKENS>
    }
}'

Response

After making an LLM cURL request, you will receive a text response. To see another example API request and response for this model, along with detailed information about the parameters, you can visit our API reference page.