A More Sustainable and Accessible Future for Palo

Since we first began sharing our vision for Mpalo, the conversations with developers, researchers, and builders have been the most rewarding part of our journey. The enthusiasm for what we are building—a true, persistent memory for AI—has been both humbling and validating. It confirms our core belief: that memory is the critical missing piece in the AI puzzle. As we move closer to our public launch, our primary responsibility shifts to ensuring the long-term health and sustainability of the platform you are excited to build on. A key part of that responsibility is creating a pricing model that is fair, transparent, and built for the long haul.

A generous free tier is a common way for platforms to attract users, but in the high-cost reality of AI computation, a one-size-fits-all free tier can be a fragile promise. It often leads to resource contention where a few heavy users can degrade the service for everyone, strict limitations that prevent meaningful work, and an unsustainable economic model that ultimately compromises the quality of the platform. In the spirit of transparency and building a platform that lasts, we are evolving our approach. Instead of a blanket free tier that serves no one well, we are focusing on a more targeted and sustainable model designed to empower the next generation of builders.

Our Core Pricing Principles

This evolution is guided by three principles that reflect our commitment to the community and our mission. They are the foundation upon which our entire business model is built.

Accessibility

We believe that cost should not be a barrier to genuine innovation. For us, accessibility is not about offering a limited free trial; it's about providing powerful, production-ready tools at the lowest possible price point. It means engineering for efficiency so that a solo developer prototyping an idea has access to the same foundational technology as an enterprise team. This principle drives our commitment to creating highly optimized, low-cost memory engines. Accessibility also means clarity. We strive to make our technology understandable and our documentation clear, ensuring that the barrier to entry is financial nor intellectual.

Transparency

You should always understand what you are paying for, without ambiguity or surprises. Our pricing is straightforward, with clear per-token costs for our engines. Your billing dashboard in the Workbench will provide real-time usage tracking and budget alerts, giving you complete control over your spending. This means you can see an estimated cost for an API call before you run it at scale and set up notifications that prevent unexpected invoices. We believe that trust is built when a user feels in control, and that begins with clear, honest pricing that eliminates the fear of a surprise bill at the end of the month.

Sustainability

A healthy platform is one that can afford to invest in its future. A pricing model that is unsustainable is a broken promise to its users. Our pricing is structured to ensure we can continue to fund our foundational research into AI memory, constantly improve our infrastructure for better performance, and maintain the reliability and security you will depend on for years to come. Sustainability isn't just about keeping the lights on; it's about having the resources to pursue the long-term research that will lead to the next generation of Palo Engines and ensure the platform you build on today becomes even more powerful tomorrow. We are building a platform to last, and that requires a sustainable economic foundation.

Our Approach: Affordable Engines for Every Project

Our commitment to accessibility is engineered directly into our products. We believe the best way to empower builders is not through a restrictive free tier, but by making our core technology fundamentally affordable. Our Palo Engine Lite and Palo Engine are not just "smaller" versions of our DEEP engine; they are highly optimized for specific use cases, allowing us to offer them at an extremely competitive price point.

As we've detailed in our post, "The Weight of Memory," this is achieved through aggressive model optimization and a focus on algorithmic efficiency. We use techniques like quantization and pruning to drastically reduce the computational cost of our lighter engines without sacrificing their core capabilities for a vast range of applications. For example, a quantized model might use 8-bit integers instead of 32-bit floats for its calculations. For a conversational AI, this has a negligible impact on the quality of the recalled memory but can reduce the computational and memory footprint by up to 75%, a saving we pass directly to you.

This means Palo Engine Lite will start at under $1 per million tokens, making it one of the most affordable ways to add powerful memory to a production chatbot, a personal prototype, or any application where speed and cost are paramount. For more demanding workloads, the Palo Engine offers a significant boost in memory capacity and contextual understanding while remaining highly cost-effective, handling more complex conversational flows and medium-term memory needs.

This tiered approach ensures you only pay for the computational power you actually need, providing a clear and affordable path to scale your applications from a simple idea to a full-fledged service, all starting from a simple, usage-based Pay-As-You-Go plan.

A Commitment to Listening

While we are not launching with a formal student or research grant program, our commitment to supporting the next generation of builders is unwavering. We understand that some of the most groundbreaking ideas come from academic labs, open-source projects, and students who are just beginning to explore the possibilities of AI. These are the environments where innovation often happens with the fewest resources, and we want to be a partner in that discovery.

The best way for us to support you is to listen.

If you are a student, an academic researcher, or a developer working on a non-commercial, open-source project and believe Mpalo can play a significant role in your work, we want to hear from you. We are always open to exploring partnerships, providing support, and finding ways to get our tools into the hands of those who are pushing the boundaries of what's possible. This could take the form of a specific grant of usage credits for a compelling open-source project, direct engineering support to help integrate Palo into a novel academic study, or simply a conversation about how we can help. Please reach out to us at mind@mpalo.com and tell us about what you're building.

The Path Forward

The pricing structure outlined here will take effect at our public launch. We believe this model—built on sustainability, transparency, and a commitment to engineering-driven affordability—is the best way to build a healthy, vibrant ecosystem for the long term. It is a system designed to grow with our users, providing a stable foundation for the amazing applications you will create. We aren't interested in being the cheapest option for a month; we are interested in being the most sustainable and valuable partner for your work over the next decade.

We are always listening. Your feedback on our pricing and platform will continue to guide our future adjustments as we grow together.