aime™ sets out its two-track direction: maximum offline reach, and a premium cloud tier for deeper educational capability

aime is pursuing two complementary tracks of work — one collapses downward to reach every classroom regardless of connectivity or hardware; the other expands upward to explore how much educational depth a modern stack can deliver when those constraints are lifted. The same operating system underpins both.

Track one — Smarter on Smaller Models, pushed to its maximum

aime is continuing to invest in the Smarter-on-Smaller-Models programme, with the explicit goal of maximising offline educational capability. The intent is straightforward: any classroom — including those without reliable internet, without modern hardware, and without on-site technical support — should be able to run curriculum-aligned, pedagogy-aware AI on what they already have. That means continuing to move work out of the model and into deterministic code, continuing to compress the agent stack onto compact education-tuned models such as aime-Reasoner-2B™ and aime-Reasoner-4B™, and continuing to harden the offline deployment surface around .aimepack™, ThinkCache™ and ThinkBook™.

• Maximum addressable footprint — low-end Windows laptops, Android tablets and iPhones.
• No regression in classroom quality as the model class shrinks.
• Offline-first by default — internet treated as an enhancement, not a prerequisite.
• Continued architectural optimisation across every agent in the stack.

Track two — premium aimeCLOUD, exploring depth without compromise

In parallel, aime is scaling aimeCLOUD to support a premium tier of educational capability for systems, universities and partners with the bandwidth and infrastructure to use it. The work here is exploratory in scope: how much further can a curriculum-aligned, pedagogy-aware system go when it can call on larger models, richer media generation and longer-running orchestration? The areas under active development include deeper technical and conceptual depth in subject reasoning, classroom-grade simulations, generated animations, gamified practice loops, and pedagogy that differentiates by student age, prior attainment and learning context.

• Deeper technical depth in subject reasoning and worked examples.
• Simulations and generated animations as native lesson primitives.
• Gamification surfaces tuned for classroom use, not for engagement-for-its-own-sake.
• Age- and stage-differentiated pedagogy, from early primary through tertiary.

“We are not choosing between reach and depth. The same operating system has to do both — collapse onto a low-end Windows laptop, an Android tablet or an iPhone for the classroom the cloud forgot, and stretch upward in aimeCLOUD to explore how far educational AI can actually go. Anything less is a half-built category.”

Summer milestones

Over the summer, aime will launch aime Hub — the ecosystem and infrastructure layer that lets ministries, universities and partners operate the platform on their own terms — alongside a programme of premium enhancements to aimeCLOUD that bring the depth-oriented capabilities described above into production.

• aime Hub launch — ecosystem and infrastructure layer for institutional operators.
• Premium aimeCLOUD enhancements — first production surfaces for simulations, animations and age-differentiated pedagogy.
• Continued Smarter-on-Smaller-Models rollout across the agent stack.

Why both tracks, at once

Educational systems are not uniform. A national curriculum has to reach classrooms with no connectivity and classrooms with research-grade infrastructure inside the same week. aime's posture is that an operating system for educational intelligence has to serve the full range — and that the design discipline required to do both at once is what produces a defensible platform, rather than another point tool.

About aime

aime builds the operating system for educational intelligence — the foundational infrastructure layer that future education systems will run on. aime's stack combines structured curriculum knowledge, pedagogy-aware reasoning, compact education-tuned models, agentic orchestration and offline-capable deployment, and is designed for ministries of education, universities and national school systems.