API Reference

Community

Blog

Try Hunch

Hunch

Support

Health

Introduction

Get your first AI block up and running in Hunch.

Your First Block

Helping AI to think and create incrementally is the best way to unlock its potential.

Connecting Blocks

Following the pattern of incremental improvements, we will now review and refine the first draft of the blog post.

Review And Refine

Enhance your blog post with an AI-generated hero image.

Image Generation

Prompts are the instructions given to AI models. Output quality depends on how well they are written for each model.

Prompting Tips

An interactive workspace for visualizing, creating, and refining AI-powered workflows

Canvas

The fundamental building blocks of AI workflows in Hunch

Blocks

Visualize data flow and dependencies in your AI workflows

Connectors

Streamline your workflow with automatic block execution

Auto-run

Intelligent model selection for optimal task performance

Auto-AI

Reusable AI workflows for enhanced productivity and collaboration

Templates

Simplified interfaces for powerful AI workflows

Tools

Share or publish your work as a reusable template, canvas, or tool for yourself, your team, or the Hunch community.

Sharing

Collaborative spaces for teams to harness AI power together

Organizations

Execute custom code within your AI workflows

Create and run your first Code Runner block in Hunch

Getting Started

Execute custom Python code within your AI workflows

Python

Execute custom Bash code within your AI workflows

Bash

Tools API

Keyboard Shortcuts

Powerful artificial intelligence engines driving your workflows

Hunch AI Model List

Reference documentation for the `hunch` CLI available in the Code Runner

Reference documentation for the `hunch` Python module available in the Code Runner

This endpoint creates a new run of the specified tool with the supplied
inputs. It returns the tool run UID, which can be used to check the status
and retrieve the outputs of the run.

Inputs are specified as a dictionary where the keys are the titles of the
input blocks and the values are the user inputs.

Start Tool Run Legacy

This endpoint returns the status of the specified tool run. If the run is
completed, it also returns the outputs of the run.

The status of the run can be one of:
- `RUNNING`: The run is still in progress.
- `COMPLETED`: The run has completed successfully.
- `FAILED`: The run has failed.

The outputs are specified as a dictionary where the keys are the titles of
the output blocks and the values are the outputs.

Get Tool Run Status Legacy

This endpoint creates a new run of the specified tool with the supplied
inputs. It returns the tool run UID, which can be used to check the status
and retrieve the outputs of the run.

Inputs are specified as a list of dictionaries each containing the title of
the input block and the input.

Start Tool Run

This endpoint returns the status of the specified tool run. If the run is
completed, it also returns the outputs of the run.

The status of the run can be one of:
- `RUNNING`: The run is still in progress.
- `COMPLETED`: The run has completed successfully.
- `FAILED`: The run has failed.

The outputs are specified as a list of dictionaries each containing the
title of the output block, the output text, and any files.