Functions

Routes incoming chat requests to a specified vision model if the user's message contains images.

Used to allow small LLMs to to make intention decisions and then dynamically switch between larger LLMs for different tasks. Also allows override based on document types. For example in the even of an image, it will use llava automatically. I will be adding user valves in a future revision to allow easier swapping of models via the UI.

A function to use flux.1.1.pro model to generate vivid images, with together.ai API key

Automatically enable Web Search and Image Generation features if the LLM requires it

A function to use flux.1.1.pro model to generate vivid images, with together.ai API key

Gemini 2.0 Flash Experimental with native image generation capabilities.

Deepseek v2.5 API both for chat and coding *** NOT SUPPORT IMAGE INPUT ***

Manifold function for Gemini Developer API. Supports native image generation, grounding with Google Search and streaming. Uses google-genai.

Allows users to generate images from text prompts with nano-gpt.com

DeepSeek-Gemini Vision integrates Deepseek with Gemini for advanced image processing. Gemini handles image analysis, while DeepSeek ensures precise results.

A simple ui for reflection. Working on image preview function.

DeepSeek-Gemini Vision integrates Deepseek with Gemini for advanced image processing. Gemini handles image analysis, while DeepSeek ensures precise results.

DOES NOT WORK WITH OWUI 0.5.X - A streamlined image generation pipeline, handling prompts, periodic status updates, and error management, designed for seamless integration with OpenWebUI's image generation tools.

Create and Edit images with gpt-image-1

Used to invoke Siliconflow's Flux model, which, when paired with the appropriate prompt, can generate corresponding images.

A function to use flux 1.1 pro ultra model to generate vivid images, with a Replicate API key.

FLUX.1 Schnell Manifold Function for Black Forest Lab Image Generation Models

A function to use Flux Schnell model to generate vivid images, with a Replicate API key.

An OpenWebUI pipe for Anthropic models that supports reasoning for the 3.7 series models and streaming, including native <thinking> tag support. Additionally, it can handle images and PDFs.

Generate uncensored images using flux.1 with novita.ai

Currently, DeepSeek V3 Pipeline API does not support image file analysis. However, this feature may be added in the future. Please stay tuned for updates from our development team.

Generates images

OpenAI Image

A simple pipe that allows you to generate images in chat using various AI models via the ImageRouter API

Generates image using Gemini 2.0 Flash with Google Genai SDK, stores into static folder and embed into chat

Manifold function for Gemini Developer API. Uses google-genai, supports thinking models, native image generation,streaming and grounding with Google Search.

Let's you set different LLM without vision support and a vision LLM that takes all Images you send and describes them to the non-vision LLM

Resize images in messages

A filter that automatically performs OCR on uploaded images using Baidu OCR API, converting image text to readable format for LLMs. 一个使用百度OCR API自动识别上传图片中文字的过滤器，将图片中的文字转换为大语言模型可读的文本格式。

Mistral API working with OpenWebui Rag and also Pixtral image

OCR image before requests, especially for o1-mini

A filter to remove markdown images in chat so they don't interfere with the context. Useful for people using their own tools and functions to generate images.

Route any attached image to describe and use its own prompt to answer

Provide enhanced cross-API visual routing capabilities to offer visual abilities for AI that does not support multimodality, facilitating the sending of images when conversing with DeepSeek-R1 and solving compatibility issues when switching models between different APIs.|提供增强的跨API视觉路由能力，为不支持多模态的AI提供视觉能力，方便与DeepSeek-R1对话时发送图片，解决不同API之间切换模型的兼容性问题

Brings OpenAI Response API support to Open WebUI. Streams <think> reasoning (if enabled) and assistant text using the OpenAI Responses SDK, with early support for images, reasoning, and status indicators.

HuggingFace API integration for Open WebUI

Filter to preprocess a message with images through a Coral TPU to augment a non-vision model.

Generate images using Venice.ai's API.

Recognize Faces In Image

This pipe function allows generating images using the official Black Forest Labs API. This function currently supports the following models: dev-1, flux-pro-1.1, and flux-pro-1.1-ultra. A raw mode switch is available for flux-pro-1.1-ultra.

This functions works for free API and Chat. Not tested with image etc.

Generates image using Stable Diffusion 3.5 API from SiliconFlow, stores into static folder and embed into chat.

This custom active button in Open WebUI allows you to modify images using the official Black Forest Labs API. This feature currently supports the following models: flux-pro-1.0-fill.

A button to generate images using Venice API. Override value feature explained: https://github.com/Simonrak/open-webui-venice

FLUX.1 Dev Manifold Function for Black Forest Lab Image Generation Model from Multi Providers with Fallback.

IONOS Image Generation

Generate images with the novita api

This action generates an image using SD models deployed on AWS Bedrock and presented via LiteLLM.

A manifold function to integrate X_AI's GROK-2-IMAGE models into Open WebUI

A dynamic filter that analyzes user input to intelligently enable relevant features like web search, image generation, or code interpretation.

Flux Replicate Image Generator for Black Forest Lab Models through Replicate (Dev, Pro, Ultra)

Anthropic enables AI-powered chat and image processing within OpenWebUI, providing access to Claude models for generating responses and handling image inputs efficiently.

Extract text from images in chat messages using OCR.Space API.

FLUX.1 Schnell Manifold Function for Black Forest Lab Image Generation Model from Several Providers.

This manifold pipe enables image generation using the official API of Black Forest Labs. Currently, only the FLUX.1.1 Pro model is supported by this function.

使用Tika OCR识别图片

Improving the standard image generation functionality

Replicate.com custom image model pipe for OpenWebUI. Includes ability to specify aspect ratio and / or number of images in the prompt. Supports working around the undocumented 4.7 megapixel limit OWUI has when processing one or more images from an image generation model.

Draw images follow the assistant's response by Siliconflow API

FLUX 1.1 Pro Manifold Function for Black Forest Lab Image Generation Model from Multiple Providers