Keep the job narrow
Extract fields, classify documents, or draft from a defined template.
Model decision / Phi
Phi earns its place on small jobs where efficiency beats raw breadth.
Phi models are designed to run efficiently on smaller hardware, which makes them useful for extraction, classification, and focused drafting on a tight footprint. We confirm the current variant, licence, and performance against the actual task.
Small footprintExtractionClassificationFocused drafting
Pattern demonstration / same workflow, synthetic material
Fixed-scope setupClient-owned accountsDraft-and-review firstFull handover
When Phi makes sense
The smallest model that passes the test wins.
Phi is a good candidate when the workflow is narrow, the hardware is modest, and a large model adds cost without adding answer quality.
Check the hardware
Confirm memory, storage, and speed for the selected Phi variant and user load.
Test the outputs
Evaluate accuracy on the team's real documents before go-live.
The Phi boundary
Efficiency is useful only when the answer is right.
Current terms win
The current licence and allowed use are confirmed before deployment.
Narrow workflows suit it
Broad, open-ended questions may need a larger or API model instead.
Human review stays final
High-stakes outputs keep a named reviewer and approval step.
A first Phi decision
Start with one narrow workflow.
Choose the task
Pick extraction, classification, or template drafting with a clear success test.
Review the model
Confirm the current variant, licence, hardware, and quality against the task.
Build the grounded setup
Connect the client-owned corpus, retrieval, and permissions.
Test and hand over
Run real documents, fix failures, and deliver the runbook and support route.
Phi is a good sign that small can be enough, when the task is small.
Why trust Pristine3D?
We build and operate production software.
Pristine3D Ltd builds and operates live digital products, and we run private AI workflows internally as part of our own operations. We scope around your real workflow: the documents you own, the questions your team asks, and the access boundary you approve. Based in Lagos, Nigeria, we work remotely with clients worldwide.
We already run live products
Based in Lagos, Nigeria, Pristine3D currently builds and operates smartcards.ng, venu.ng, photoshoot.ng, and ugc.ng in production.
We start with the actual workflow
One input, one output, one test set, and one person who owns the result. We scope a real workflow instead of a transformation programme.
The boundary stays visible
Cloud, model, storage, and messaging accounts stay in your name. The chosen data path, access rules, test record, documentation, and training are part of the agreed scope.
Pricing / fixed scope
Know the starting numbers before you ask.
The final quote follows the workflow. Infrastructure and model bills stay on your accounts.
One Workflow
Starting from$3,500 / ₦2.5m
2 to 3 weeksOne workflow for a small team, with training and handover.
See the one-workflow packageFirm-wide
Starting from$7,500 / ₦8m
4 to 6 weeksMultiple sources, roles, integrations, and admin handover.
See the deploymentAnnual support
Starting from$3,000 / ₦1.5m
per yearStandard care for one delivered workflow. Optional. Larger deployments and active monitoring are separately scoped.
See supportArchitecture review from $500. Standard annual support is $3,000 / ₦1.5m per year for one delivered workflow. New workflows, integrations, active monitoring, and infrastructure are separately scoped; infrastructure, model, storage, and messaging bills stay on client accounts. Full pricing and what changes the quote
Straight answers
Common questions.
Can Phi handle legal or financial documents?
It can extract and classify when the workflow and evaluation set are scoped correctly. The professional still owns review and sign-off.
Does it need a GPU?
Many Phi variants run efficiently on modest hardware. Exact requirements are confirmed for the chosen variant and workload.
What if the task grows later?
The workflow can move to a larger model or an API route without rebuilding the knowledge base.
Own your knowledge base
The model is not the product. The knowledge base is.
Documents, the retrieval index, access rules, and the workflows built around them are the asset, and they compound. We deploy so the knowledge base stays yours: on your accounts, in the environment you choose, under access rules your team defines. The model behind the answers is a connector, so the knowledge base moves with you, not with a vendor.
Your corpus, your index
The document store, metadata, and retrieval setup live on accounts you own. No vendor holds the corpus.
Models are swappable parts
Change the model provider, move regions, or go local without rebuilding the knowledge base or the workflow.
The knowledge base is the alpha
Every improvement to the corpus improves the answers, and the improvement stays with you, not with a vendor.
Keep exploring
Related setups.
Start with the task
Which narrow job should run on a small model?
Tell us the extraction, classification, or drafting task and the hardware constraint. We will tell you if Phi fits.