RFPs: Open Source AI Model for Tutoring (US)

Deadline: 31-Jul-2026

The Open Source AI Model for Tutoring Request for Proposals (RFP) provides funding to develop open-source artificial intelligence models designed to deliver effective K–12 mathematics tutoring in the United States. The initiative supports the creation of AI tutoring systems that can provide personalised learning experiences, improve student outcomes, and achieve effectiveness comparable to expert human tutors.

The grant supports research, model development, educational AI infrastructure, and collaboration between technology organisations, researchers, educators, and other stakeholders to advance open-source AI solutions for mathematics education.

Overview of the Open Source AI Model for Tutoring Grant

The Open Source AI Model for Tutoring initiative supports the development of education-focused AI systems that improve mathematics learning for students in K–12 schools across the United States.

The programme focuses on building open-source AI models that can:

• Provide personalised one-to-one mathematics tutoring.
• Support independent student learning.
• Improve student engagement and motivation.
• Strengthen mathematical understanding and learning outcomes.
• Align with teacher instructional methods and classroom approaches.
• Use multiple forms of educational content, including text, audio, video, and student-created drawings.

The programme is designed to advance AI technologies that can operate effectively within real educational environments.

Purpose of the Open Source AI Model for Tutoring Grant

The primary purpose of the grant is to support the creation of AI tutoring systems that can provide learning support similar to human expert tutors.

The initiative aims to:

• Improve the quality and accessibility of mathematics education.
• Develop open-source educational AI models.
• Advance research in AI-supported learning.
• Create tools that support teachers and students.
• Encourage collaboration within the open-source AI education ecosystem.
• Strengthen understanding of how AI can improve learning.

The programme also supports research activities that help improve educational AI capabilities in structured learning environments.

What Is AI Math Tutoring?

AI math tutoring refers to a one-to-one interaction between a student and an artificial intelligence system within a K–12 educational setting.

An effective AI math tutor should:

• Help students improve mathematics learning outcomes.
• Provide personalised guidance.
• Adapt to individual student needs.
• Support independent learning.
• Follow educational approaches aligned with teacher instruction.
• Provide explanations, feedback, and learning support.

The system should function as an educational partner rather than simply provide answers.

Key Focus Areas of the Grant

The Open Source AI Model for Tutoring grant supports projects across several important areas.

Open-Source Education AI Model Development

Projects should focus on developing AI models specifically designed for educational use.

Supported activities may include:

• Building open-source large language models.
• Creating education-specific AI systems.
• Improving model reasoning and tutoring abilities.
• Developing reusable AI resources for education.

AI-Based Mathematics Tutoring Systems

Projects should develop AI systems capable of supporting mathematics learning.

Systems should help students:

• Understand mathematical concepts.
• Solve problems independently.
• Receive personalised feedback.
• Develop stronger problem-solving skills.

K–12 Learning Environments in the United States

AI models should be designed for use within U.S. K–12 education settings.

Applicants should consider:

• Classroom realities.
• Teacher workflows.
• Student learning needs.
• Curriculum requirements.
• Educational accessibility.

Student Motivation and Engagement

Projects should explore ways AI can encourage students to remain engaged in learning.

Possible approaches include:

• Personalised encouragement.
• Interactive learning experiences.
• Adaptive challenges.
• Feedback systems that support confidence.

Metacognition and Learning Improvement

Effective AI tutors should help students understand how they learn.

Projects may support:

• Reflection on problem-solving approaches.
• Self-assessment skills.
• Learning strategies.
• Improved mathematical reasoning.

Teacher Instructional Alignment

AI systems should work alongside teachers by aligning with:

• Teaching methods.
• Curriculum approaches.
• Classroom learning goals.
• Instructional plans.

The goal is to support educators rather than replace them.

Multimodal Learning Resources

Projects should explore AI systems that can work with multiple learning formats.

Supported resources may include:

• Text explanations.
• Audio guidance.
• Video learning materials.
• Student drawings.
• Visual learning tools.

Multimodal capabilities can help AI tutors better understand and support different learning styles.

Research and AI Education Infrastructure

Funding may support:

• Educational AI research.
• Model evaluation systems.
• Training infrastructure.
• Testing environments.
• Open-source development resources.

Stakeholder Feedback Integration

Successful projects should include feedback from:

• Teachers.
• Students.
• Education researchers.
• School communities.
• Other educational stakeholders.

Feedback should inform model development and improvement.

Funding Amount Available

The grant provides:

• Up to $8,000,000 USD in funding.
• An estimated project period of 30 to 36 months.

Funding is intended to support both AI model development and research activities.

Projects should demonstrate how funding will contribute to scalable, open-source educational AI solutions.

Who Is Eligible to Apply?

Eligible applicants include organisations and institutions with demonstrated experience in artificial intelligence model development for education.

Applicants may include:

• Technology organisations.
• Research institutions.
• Educational technology organisations.
• Academic institutions.
• Organisations developing open-source AI resources.

Applicants must demonstrate the technical and educational expertise required to successfully deliver the project.

Applicant Requirements

Applicants must show evidence of:

AI Model Development Experience

Applicants should have:

• Experience developing large language models.
• Knowledge of AI system design.
• Experience creating educational AI applications.

Open-Source Contribution Experience

Applicants should demonstrate contributions to digital public goods, such as:

• Open-source AI models.
• Public datasets.
• Evaluation tools.
• Shared research resources.

Real-World Model Deployment Experience

Applicants must demonstrate:

• Deployment or evaluation of AI models using real student or user data.
• Experience operating systems at meaningful scale.
• Ability to test models in practical learning environments.

Educational Data and Evaluation Capability

Proposals must include plans for:

• Training AI models using U.S. K–12 educational data.
• Testing system performance.
• Measuring student learning outcomes.
• Evaluating educational effectiveness.

Stakeholder Engagement Plans

Applicants must explain how they will gather input from:

• Teachers.
• Students.
• Education experts.
• Other relevant stakeholders.

This feedback should guide system development and improvement.

Partnerships and Collaboration Requirements

Partnerships between organisations are encouraged and may be necessary for applicants to meet all programme requirements.

Collaborations may combine expertise in:

• Artificial intelligence.
• Education research.
• Classroom implementation.
• Data management.
• Open-source development.

Strong partnerships can improve the ability to create effective AI tutoring systems.

Projects That Are Not Eligible

The grant does not support proposals focused only on standalone applications or limited technology solutions.

Projects will not be considered if they are:

• Simple AI applications without underlying model development.
• Point solutions addressing only one narrow problem.
• Closed systems without open-source contributions.
• Tools without educational research components.

The programme prioritises foundational AI model development and ecosystem advancement.

How the Open Source AI Model for Tutoring Grant Works

Applicants should follow these steps when preparing proposals:

Step 1: Define the AI Education Challenge

Applicants should identify:

• The mathematics learning challenge being addressed.
• The target student population.
• The educational environment.
• The expected learning improvements.

Step 2: Develop the AI Model Strategy

Proposals should explain:

• Model development approach.
• Open-source strategy.
• Training methods.
• Evaluation plans.
• Educational alignment.

Step 3: Build Stakeholder Partnerships

Applicants should identify:

• Education partners.
• Researchers.
• Teachers.
• Schools or learning communities.

Partnerships should support effective testing and improvement.

Step 4: Submit a Research and Implementation Plan

Proposals should include:

• Project goals.
• Development timeline.
• Research methods.
• Budget requirements.
• Expected outcomes.

Step 5: Develop, Test, and Improve the AI System

Funded projects should:

• Build the AI tutoring model.
• Test performance with real users.
• Collect stakeholder feedback.
• Improve system effectiveness.
• Share open-source resources.

Why the Open Source AI Model for Tutoring Grant Matters

Artificial intelligence has the potential to expand access to high-quality learning support, especially in mathematics education.

The programme supports:

• More personalised student learning.
• Greater access to tutoring resources.
• Improved educational research.
• Stronger open-source AI development.
• Better collaboration between technology and education sectors.

By investing in open educational AI models, the programme aims to improve the future of mathematics learning.

Benefits for Students, Teachers, and Education Systems

Students may benefit through:

• Personalised mathematics support.
• Increased confidence.
• More opportunities for practice.
• Improved learning outcomes.

Teachers may benefit through:

• AI-assisted instructional support.
• Better understanding of student needs.
• Additional learning resources.

Education systems may benefit through:

• Open-source educational technology.
• Research-driven AI tools.
• Scalable learning solutions.

Tips for Preparing a Strong Proposal

Applicants can strengthen their proposals by:

• Demonstrating previous AI model development success.
• Providing evidence of educational impact.
• Explaining open-source commitments clearly.
• Including teacher and student feedback strategies.
• Showing experience with large-scale deployment.
• Building strong partnerships.
• Defining measurable learning outcomes.

Common Application Mistakes to Avoid

Applicants should avoid:

• Proposing only a standalone tutoring application.
• Failing to explain open-source contributions.
• Ignoring teacher involvement.
• Using limited or unrealistic evaluation methods.
• Not demonstrating experience with large language models.
• Creating technology without a clear educational purpose.

Frequently Asked Questions

What is the Open Source AI Model for Tutoring grant?

The grant supports the development of open-source AI models that provide effective mathematics tutoring for K–12 students in the United States.

How much funding is available?

Projects can receive up to $8,000,000 USD.

Who can apply for the grant?

Organisations and institutions with experience in AI model development for education, large language models, and open-source digital resources may apply.

What type of AI systems does the grant support?

The programme supports AI tutoring systems that provide one-to-one mathematics learning support, personalised feedback, and educational guidance.

Are standalone AI tutoring apps eligible?

No. Proposals focused only on standalone applications or point solutions will not be considered.

Do projects need to use real student data?

Yes. Applicants must demonstrate plans for training and testing models using U.S. K–12 educational data and evaluating performance with meaningful user data.

Are partnerships encouraged?

Yes. Partnerships are encouraged and may be required to combine technical, educational, and research expertise.

Conclusion

The Open Source AI Model for Tutoring grant provides major support for developing next-generation AI-powered mathematics tutoring systems for K–12 education in the United States. By funding open-source AI models, educational research, and real-world testing, the programme aims to create scalable tools that improve student learning and support teachers.

Successful applicants should demonstrate strong AI expertise, educational understanding, open-source commitment, and the ability to create AI tutoring systems that deliver meaningful benefits for students and schools.

For more information, visit K12 AI Infrastructure Program.