OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Instructir

InstructIR is an MIT-licensed computer-vision model and research project that uses a human-written instruction alongside an image to guide restoration. It addresses tasks including denoising, deraining, deblurring, dehazing and low-light image enhancement.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Instructir — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use InstructIR when you want to explore image restoration guided by a written instruction, particularly for the restoration tasks listed by the project.

Who it is for

  • People exploring restoration of degraded or low-light images through natural-language requests.
  • Researchers and developers experimenting with instruction-guided image restoration.
  • Users evaluating the project through its hosted demonstration or Colab tutorial.

Documented uses

  • Direct an image-restoration model with a specific written request, such as removing raindrops while preserving image content.
  • Explore denoising, deraining, deblurring, dehazing or low-light enhancement.
  • Try the project through its hosted demo or notebook tutorial, or prepare the documented local Gradio demo environment.
Before you choose

Requirements & limits

The repository README describes the Hugging Face demo as available at no cost and without code. It also links to a Colab tutorial and describes a local Gradio demo requiring Python 3.10 or later and the listed requirements.

  • The model uses both an input image and a human-written instruction, rather than describing a single fixed restoration operation.
  • The repository documents hosted, notebook-based and local Gradio routes to explore the project.
  • The local Gradio demo has a documented Python 3.10-or-later prerequisite and a requirements file.
  • The repository characterizes InstructIR as a research project, not a product.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare the local Gradio demo environment

Set up the documented prerequisites for running InstructIR’s local Gradio demo.

Before you start

  • Python 3.10 or later
  • The requirements listed in the repository’s requirements_gradio.txt file

Steps

  1. Open the InstructIR repository and locate gradio_demo.py and requirements_gradio.txt.

  2. Install Python 3.10 or later if needed.

  3. Install the requirements listed in requirements_gradio.txt.

Expected outcome

The documented Python and package prerequisites for the local Gradio demo are in place.

Guide limitations

  • The README identifies the demo and its prerequisites but these steps do not establish that the demo has been launched or that a restoration completed.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
mv-lab/InstructIR · Official homepage title · retained in its source language
How we review sourcesSuggest a correction