Nzitsa AI Labs and Nzeru
A functional AI platform I built around one governed API for model inference, with hundreds of OpenRouter models connectable for free and paid customer offerings; Studio and GPU services continue to evolve.
Building capability, not only access
Nzitsa AI Labs began with a practical question: how can more Zambians do more than use global AI tools? A person can open a general chatbot in seconds, but turning a local idea into a useful, safe product still takes access decisions, model selection, cost controls, technical review, and people who know how to build it.
AI Labs is the functional platform layer for that work. Its provider gateway presents a single Nzitsa API endpoint while controlling which models an account may use, how requests are routed, what usage costs and how activity is reviewed. Hundreds of OpenRouter models can be connected, spanning free and paid offerings. That describes the breadth of possible model connections; each customer's available models depend on account permissions, reviewed routes and pricing. Hosted providers supply inference; Nzitsa provides the customer-facing API, access rules, routing and operating controls.
Nzeru and the route into practical work
Nzitsa Nzeru is the people side of the initiative. Through Know About AI seminars, mentorship and practical project pathways, it helps learners, educators, researchers and early-career builders grow from curiosity into contribution. The first seminar, Artificial Intelligence for All Zambians, is scheduled for 21 November 2026. Approved builders can use the governed AI Labs environment and participate in reviewed work.
The K80,000 EIZ Engineering Fund award announced on 10 April 2026 gives this work important momentum. It supports the AI Labs and Nzeru direction, with remaining grant funding intended for the Know About AI seminar programme.
Nzitsa Studio and the next layer
Nzitsa Studio is the next layer being perfected: a visual flow where a builder can begin with a reviewed chat template, state an assistant's purpose, choose an account-available model, adjust the interface and preview the result on desktop and mobile. GPU-backed training and compute are also active development areas. These services have different technical and cost controls from routing an inference request to an existing hosted model.
The public AI Labs site gives visitors an AI product gallery, developer entry point, Nzeru Build Space and seminar links. The inference gateway and single API are functional. The source application also contains collaboration, dataset and Studio workflows; GPU, training, deployment, compute-credit and model-hosting offers continue to be refined and released under their own controls.
My role is to connect that infrastructure to useful products and a route for local builders to participate. Nzeru is the community and learning programme; AI Labs is the product and technical platform. Read the seminar update or the fuller Block Studio founder's note, visit Nzitsa AI Labs, or explore Nzitsa Platforms.
Local governance, hosted capability
AI Labs manages identity, roles, model access, routing, quotas, cost records, approvals and audit trails around a single customer-facing inference API. Hosted providers supply the underlying models.
Model reach
Hundreds of OpenRouter models can be connected to the platform and offered through one Nzitsa API endpoint, with free and paid options. Actual customer access depends on reviewed routes, account permissions and pricing controls.
What visitors can see
The public AI Labs site introduces a product gallery, learning tools, developer access, Nzeru Build Space and seminars. The platform and inference gateway are functional, while individual products and infrastructure services have their own availability rules.
Nzitsa Studio and GPU work
Nzitsa Studio is being refined as a visual building experience with reviewed templates and previews. GPU-backed training and compute services are also being perfected; they should not be confused with the already functional inference gateway.
Nzeru develops the people
Nzitsa Nzeru builds local confidence and capability through Know About AI seminars, mentorship, and practical pathways. AI Labs then gives approved builders governed tools and opportunities to contribute to reviewed projects.
Current boundary
The single API inference platform works. This page does not claim that all connectable models are already active for every account or that Studio, GPU training, compute credits and model hosting are finished public services.