Platform

How the Idirak platform works

The ecosystem, featured products, AI models, demo videos and getting-started steps, all on one page.

Browse all tools
Idirak V2 · Connected ecosystem

One intelligent home for Uyghur AI services

Discover language, speech, OCR, education, creativity, and everyday tools through one coherent Idirak experience.

This view organizes related services; it does not imply that every tool exchanges data with every other tool.

One engine. Every idirak tool.

  • check

    One engine, five tools

    Notes, AI OCR, text-to-speech, the Yardem agent builder, and AI chat all run on the same Uyghur-language engine — what you learn in one tool carries over to the rest.

  • check

    Purpose-built for Uyghur

  • check

    Always expanding

Browse all tools

Notes, AI OCR, text-to-speech, the Yardem agent builder, and AI chat all run on the same Uyghur-language engine — what you learn in one tool carries over to the rest.

Idirak

Tap a tool to read what it does

Creative intelligence

AI built for creators

Reveal the hidden power of AI-enabled creative tools.

Idirak Intelligence

Creative AI system · V2.0

System online
Idirak AI creative assistant
AI vision systemContext synchronized98.6%
Multimodal creative engine

AI assistant

Powerful resources that work effortlessly for you.

  • Image generation01
  • Image enhancement02
  • Image compositing03
LoadingAI in progress|
Original blurry photo
AI-enhanced sharp result
01 / VISION

Image enhancement

Sharpen blurry shots into crisp, high-resolution visuals in one click.

Enhance this photo
AI motion system02 / MOTION

Video generator

Hollywood-level editing tools let you craft cinematic stories at home.

Video render complete
Brainwave

Just now

AI Demo Videos

Explore the powerful capabilities of artificial intelligence

AI Advanced Technology

Smart System

Future AI World

Emerging AI

AI Innovation

Intelligent World

Idirak Model Lab

A future model family for Uyghur AI

Kiroran, Lopnur, Tarim, and Taklimakan are proposed model directions for different use cases, all sharing an Uyghur-first foundation.

Unreleased model concepts

The names and target profiles below communicate a future direction. They are not released models, public APIs, benchmark results, or guaranteed technical specifications.

Concept / RoadmapK-01

Kiroran

General reasoning model

A proposed flagship model for everyday assistance, structured reasoning, planning, coding, and tool-oriented work.

Speed target
Target: balanced
Context target
Planned: extended
Modalities
Text · Image · Code
Language direction
Uyghur-first · Multilingual

Target capabilities

  • Everyday and professional assistance
  • Structured reasoning and planning
  • Coding, tools, and task coordination

No public benchmark or production API is available yet.

Concept / RoadmapL-02

Lopnur

Compact and fast model

A compact model direction for responsive conversations, voice assistance, and efficient everyday tasks.

Speed target
Target: low latency
Context target
Planned: compact
Modalities
Text · Speech
Language direction
Uyghur-first

Target capabilities

  • Responsive chat and voice assistance
  • Efficiency-oriented design
  • Short everyday task support

No public benchmark or production API is available yet.

Concept / RoadmapT-03

Tarim

Knowledge and long-context model

A knowledge-work model concept for long documents, research, retrieval, and source-grounded responses.

Speed target
Target: deliberate
Context target
Planned: long-form
Modalities
Text · Documents · Retrieval
Language direction
Uyghur-first · Multilingual

Target capabilities

  • Long-document analysis
  • Source-grounded search and answers
  • Research and knowledge workflows

No public benchmark or production API is available yet.

Concept / RoadmapTK-04

Taklimakan

Creative multimodal model

A creative model direction for cultural content, education, storytelling, and multimodal workflows.

Speed target
Target: interactive
Context target
Planned: project-scale
Modalities
Text · Image · Audio
Language direction
Uyghur and cultural context first

Target capabilities

  • Cultural content and storytelling
  • Image, audio, and text workflows
  • Education and creative projects

No public benchmark or production API is available yet.

Actual capabilities and technical specifications will only be defined after research, evaluation, and transparent testing are complete.

Ready to get started?

Your next idea starts here.

Choose a tool and turn your words, documents, or ideas into something useful. Start small and make it yours.

Three simple steps

  1. Choose your tool

    Find the right tool for what you want to do.

  2. Bring your content

    Start with your text, image, or idea in the tool you choose.

  3. Make it yours

    Review the result, refine it, and put it to work in your next project.