$47 a month.
Every line item public.
Sovereign Superintelligence at one-hundredth the cost of the frontier-lab stack. Below is the actual invoice for one production AlifZetta node serving ninety-four-thousand entries at sub-5ms latency.
The real invoice — this month
| Line | Amount | What it bought |
|---|---|---|
| Compute (Azure D8s v5, monthly) | $34.60 | 1 CPU node running our full stack, 24/7 |
| Storage (256 GB Premium SSD) | $7.80 | NEXUS substrate + logs + cache |
| Bandwidth (egress ~40 GB) | $3.50 | Public API traffic |
| TLS cert (LetsEncrypt) | $0.00 | Free |
| Vector DB SaaS | $0.00 | We do not use one |
| Frontier LLM API | $0.00 | We do not use one |
| GPU rental | $0.00 | We do not need one |
| Embedding API | $0.00 | We do not use embeddings |
| Total | $45.90 | Rounded to $47 for the headline |
What that $47 actually serves
Ninety-four thousand cited NEXUS entries · sub-5ms retrieval · bilingual English + Nepali · sub-3ms partner-API latency · 92.7% cache hit rate · zero external LLM calls · full PRISM proof-tree on every answer. Live at demo.axz.si right now.
Pilot & production tiers
| Dimension | Self-hosted (free) | Two-week pilot | Production partnership |
|---|---|---|---|
| Deployment | Your infrastructure | Your infrastructure, we install | Your infrastructure, we operate |
| Software | Open code — MIT/Apache | Full stack pre-configured | Full stack + SLA + updates |
| Support | Community | Founder-direct (Padam) | Priority + on-call |
| Ingest your knowledge | DIY (docs provided) | We do it — up to 100k entries | We do it — unlimited |
| Custom vertical KB | — | One vertical (healthcare, gov, etc.) | Any + on-demand |
| Bilingual (EN + NE / HI) | ✓ | ✓ | ✓ + any additional language |
| Response SLA | Best-effort | 48h founder response | 4h business-hours · 24h off-hours |
| Auditor sign-off support | — | Included | Included + regulatory letters |
| Infrastructure cost/mo | ~$47/mo (yours) | ~$47/mo (yours) | ~$47/mo (yours) |
| Our fee | $0 | Fixed-fee pilot (email for quote) | Retainer + success bonus |
What we do not charge for
Per-token fees. Per-query fees. Per-active-user fees. Vector-DB seats. Embedding calls. Retriever tuning. LLM-judge overhead. Third-party API pass-through markup. Model version upgrades. Language add-ons. We keep the pricing shape as simple as the architecture — because that is the point.
Frequently asked
Why so cheap? Because the architecture is CPU-first and citation-fenced. No GPU rent. No vector-DB SaaS. No frontier-lab API. The paradigm's whole point is to flip the cost curve — and it does.
What if we need 10x more throughput? Add another Ryzen node. Load balance. Cost scales linearly to about $500/mo for a 10x rack — versus tens of thousands with the GPU stack.
Do you offer a free trial? The whole live demo at demo.axz.si is the free trial. Sub-5ms cited retrieval on ninety-four-thousand entries. Query anything.
What is included in the two-week fixed-fee pilot? We install the stack on your hardware, ingest up to a hundred-thousand of your entries into NEXUS, build one vertical CLLM (healthcare, government, finance, whichever fits), and hand you a working authenticated endpoint with a monitored baseline. Email padam@axz.si for a quote.
Book a pilot conversation
30 minutes. No slides. We map your workflow to the AlifZetta stack live, quote the pilot on the call.
Email padam@axz.si → Meet the founder →