EduSlopTutor

Paper-Mediated AI Tutor

Start a New Education Program

Create a learning program by defining the subject, level, and goals. We'll generate a knowledge bag (reference curriculum) that you can review before generating personalised worksheets for your students.

1. Configure
2. Generate Knowledge
3. Review
📁

Click or drag files here

YAML config, reference images, or supporting materials

How the new flow works

  • Step 1: Create a program — AI generates a comprehensive knowledge bag (reference curriculum)
  • Step 2: Review the knowledge bag in 📚 My Programs
  • Step 3: Select a class and generate personalised worksheets for each student
  • Each student gets worksheets tailored to their current knowledge level

Submit Completed Homework

Upload a photo or scan of your completed worksheet. The AI grading system will analyze your answers and provide feedback.

1. Upload
2. Processing
3. Results
📷

Click or drag your worksheet photo

PNG, JPG, or PDF — clear photo of the completed worksheet page

Tips for best results

  • Take a clear, well-lit photo of the entire worksheet page
  • Keep the image straight and avoid shadows
  • Make sure handwriting is legible
  • The QR code on the page links directly here with the lesson ID pre-filled

📚 My Programs

Explore your programs, review knowledge bags, generate worksheets for a class, and track student progress.

💰 Cost Breakdown

View the AI generation costs for each of your projects.

📊 Platform Overview

System-wide stats across all teachers.

👨‍🏫 Teachers

Per-teacher usage and costs.

Teacher Email Projects Lessons Students Cost
Loading…

🎓 Student Roster

Manage students organised by class.

⚙️ AI Model Settings

Choose which AI models to use for each step of the generation pipeline. Changes are saved locally and applied to all new jobs.

Used for generating hero images on the front page.

Used for generating the subject knowledge bag (topics, concepts).

Used for generating lesson content, sections, and questions.

Used after grading to assess student knowledge progress. Requires structured JSON output.

Model Guide

  • GPT-4o Mini — Fast and cheap, strict JSON schema — ideal default for lessons
  • GPT-4o — Reliable workhorse, strict JSON schema + strong vision
  • GPT-5.2 Chat — Most capable reasoning, strict JSON schema — best for knowledge aggregation
  • GPT-5 Mini — Good reasoning, cost-effective, works with structured output
  • DeepSeek V3 Speciale — Strict JSON schema, very cheap, but low throughput (3 RPM)
  • Phi-4 — Microsoft’s free model — knowledge generation only (no strict JSON schema)
  • DALL·E 3 — Image generation with good prompt following

Lesson generation and knowledge assessment both require structured JSON output. Phi-4 is excluded from those lists as it does not support strict JSON schema.

Question crop