pollinations/pollinations

Your Friendly Open-Source Gen-AI Platform

What it solves

Pollinations provides a unified, open-source API for generative AI, removing the need to integrate multiple different services for text, image, audio, video, and 3D generation. It simplifies the process for developers to access a wide variety of models (like Flux, GPT, Claude, and Gemini) through a single endpoint and a consistent authentication system.

How it works

The platform operates as a unified endpoint (gen.pollinations.ai) that routes requests to various AI models. It offers an OpenAI-compatible API, allowing developers to use existing OpenAI SDKs or the official Pollinations SDK. Users can manage their usage via "Pollen" credits and authenticate using either public App Keys (pk_) or server-side Secret Keys (sk_). Additionally, it supports "Managed Agents" where users can define a system prompt and base model to create reusable AI agents without hosting their own server.

Who it’s for

It is designed for developers, creators, and AI agents who want to access multimodal generative AI capabilities without the overhead of managing multiple API keys and infrastructure. It also supports community members who wish to host and register their own OpenAI-compatible models to the platform's catalog.

Highlights

  • Multimodal Generation: Supports text, image, audio, video, 3D, and embeddings.
  • OpenAI Compatibility: Works out-of-the-box with official OpenAI SDKs and the Vercel AI SDK.
  • Managed Agents: Create and publish reusable AI agents with custom system prompts and tools.
  • Community Model Hosting: Allows users to connect their own endpoints to the Pollinations catalog with automatic fallback routing.
  • Developer Tooling: Includes a CLI for terminal-based generation and an MCP server for integration with AI assistants like Claude.

Related

  • Project
    flatkey-ai/flatkey-cli

    A command-line interface for unified multimodal AI generation, allowing media teams and AI agents to generate images, videos, audio, and text through a single API key and credit balance.

  • Project
    Anil-matcha/Open-Generative-AI

    An unrestricted, open-source AI studio for generating images, videos, and audio using 400+ models without content filters or subscription fees.

  • Project
    mcmonkeyprojects/SwarmUI

    A modular AI generation web interface that supports image, video, and audio models and can utilize multiple GPUs simultaneously for high-performance generation.

  • Project
    AgnesAI-Labs/AgnesAI-Models

    A unified OpenAI-compatible API gateway providing access to high-performance multimodal foundation models for text, image, and video generation.

  • Project
    GT4SD/gt4sd-core

    An open-source platform that provides a unified API and toolkit for using state-of-the-art generative AI models to accelerate hypothesis generation in scientific discovery.