Sovereign AI. Governed. Custom-orchestrated.
Product 1 · Salient 1.0
Everyone is building a bigger brain. We built the suit.
Salient 1.0 is our own AI model: a 27B open-weight base, post-trained in-house, then frozen. Twelve engineered layers around it retrieve, verify, execute and decide. It runs on one graphics card, keeps working at 5M tokens, and hands you a receipt for every claim.
Brain. Frozen 27B, multimodal, one GPU.
Scored on our own Harness: tests we built and run ourselves, which you are welcome to rerun.
The powersuit
Salient 1.0The power lives in the suit, not the wearer.
Most AI gets better by getting bigger. Salient gets better by being better equipped. The model at its centre is frozen once it is deployed, so it never quietly changes and you can always explain what it did.
The capability comes from twelve engineered layers around it. And because the power lives in the suit, a better model is a configuration change, not a rebuild. When the next open-weight model arrives, Salient re-fits to it overnight, and every pack, every receipt and every verified line of code carries straight over.
- 01BrainFrozen 27B, multimodal, one GPU
- 02LibrarySigned, versioned knowledge packsCorpus
- 03LibrarianHybrid search + reranking, 2.95× faster
- 04CoachEffort that scales with difficulty
- 05Fact-CheckerByte-for-byte quote verification
- 06Bouncer & ReceiptServer-side gate, tamper-evident provenance
- 07MemorySearchable past 64K, works at 5M
- 08Freshness RingDaily ingest, isolated regulated laneCorpus
- 09Coder’s Toolbox3,556 tested routines + sandbox
- 10SwitchboardRoutes general, verified and code
- 11Control RoomFull provenance, self-tuning optimisers
- 12Decision ModeCalibrated confidence in milliseconds
Layers 1, 3–7 and 9–12 are the engine: model, retrieval, verification, execution, memory, decision-making. They don’t change by geography. Layers 2 and 8 are the corpus: India’s statutory and regulatory material ships today. A new country’s statutes, or your company’s own domain, is the same ingestion pipeline pointed at a new source. Weeks, not years.
Context engineering
Salient 1.0Past 64,000 tokens, Salient stops straining and starts searching.
Conventional context windows have a ceiling. Salient’s memory turns a long conversation into something it searches, with hybrid search, precision reranking and the evidence kept in document order. Drag the slider to see what happens as the conversation grows.
Searched, not strained
Beyond the live window, Salient searches what it has read (hybrid search, precision reranking, evidence kept in document order) instead of straining to remember it. It keeps working at 300K, 1M and 5M tokens, where conventional context windows stop functioning.
180K
tokens of live context on a single graphics card
5M
tokens of unchunked context
100B
tokens in the indexed library
2.95×
faster retrieval, and more accurate, not less
Proof, not guesswork
Salient 1.0The AI that always shows its proof.
Beats leading AI models on India
Scores 20.8 points higher than a leading AI model on Indian tax and regulatory questions. That’s not a bigger model — it’s knowledge we built ourselves.
Proof, not guesswork
Every quote is checked word-for-word against its source before it’s shown to you. We tried six times to sneak a fake quote past this check. None got through.
A score, not a guess
The AI gives a confidence score with every answer, in a fraction of a second. If it isn’t confident, it says so or asks for a person — it never just guesses.
Ours, not rented
We trained it ourselves and we own it outright — the model, the training process and the roadmap. No dependence on a foreign company’s pricing or access.
Runs on one graphics card
The entire system runs on a single high-end graphics card — the kind built for serious AI work — in your own office. No need to build a data centre or hire a specialist infrastructure team.
Independently tested
Every number on this page comes from tests we built and run ourselves — not someone else’s leaderboard. You’re welcome to test it yourself, too.
Every claim is scored on our own Harness: evaluation engineering we built in-house, not a borrowed leaderboard. You’re welcome to rerun it against your own tests.
- Pick one costly, repetitive workflow — not a whole department.
- Point Salient at your historical cases and outcomes, and run it in parallel with your team on our Harness.
- Measure the delta in accuracy, turnaround time and cost against your baseline. You keep the results either way.
The Provenance Stack
Three productsA brain, a governor and a control room.
Salient reasons. Umpire checks every action before it happens. Verity puts both to work across the apps you already use, from WhatsApp and Gmail to Slack, Excel and your databases. It is AI you own outright, not AI you rent by the token.
Verity
Ask in plain English. It pulls what’s needed from the apps you already run and builds the answer: a financial model, a follow-up message, a status report. It doesn’t do the thinking itself.
Salient 1.0
Our own AI model, trained in-house, running on a single graphics card. Always shows where its answer came from.
Umpire
Checks every action before it happens, and fixes it if a rule was ever wrong. Works on any agent, ours or a vendor’s.
- Gmail
- Company email
- Slack
- Excel
- CRM
- Calendar
- Call recordings
- Ticketing
- Databases
Product 2
Umpire
AI agents your auditor will accept
AI systems today can approve payments, change records and write code, often without a person checking each step. Most tools only check what an AI says. Umpire checks what it does.
It sits between any AI agent, ours or anyone else’s, and the systems it touches. Nothing important happens without a check, one person can stop everything instantly, and every action leaves behind a record that can’t be quietly changed. And if a rule turns out to have been wrong, Umpire finds every decision it affected and fixes them.
Checks before it happens
Each AI agent is given a clear purpose, a time limit and a spending limit. Anything not specifically allowed is blocked before it can touch a real system. Agents never get your passwords.
One-button stop
An authorised person can pause one AI agent, or all of them, instantly. Two people are needed to restart. Records keep being kept even while paused, so the pause itself is provable.
Fixes what went wrong
If a rule was wrong for eight months, one request finds every decision it affected and drafts the fix. Work that used to take consultants months now takes an afternoon.
A clean actionSettle an approved claim with the registered payee.
- 01PrepareAgent declares the exact action and payload
- 02Mandate checkDeterministic evaluation — scope, value, recipient, rate, window, content
- 03TokenSingle-use, bound to the exact payload hash, expires in seconds
- 04SubmitPayload re-hashed and compared to the token binding
- 05Execute / blockRuns once, or stops — a timeout is never permission
- 06Seal receiptHash-chained to the receipt before it
Every consequential action goes through the same six steps. Press a button to run one.
Umpire’s AI firewall is included on every Salient deployment.
Product 3
Verity
One place to run everything — with proof it worked
Ask Verity, in plain English, to pull what you need from WhatsApp, Gmail, company email, Slack, Excel and any other software or database you use, and it builds the answer for you: a financial model, a follow-up message, a status report, put together from apps that were never designed to talk to each other.
Verity doesn’t do the thinking itself. It puts Salient’s reasoning and Umpire’s checks to work across every app your company already uses, and every action it takes is permanently recorded, so it can be checked years later.
One place for everything
Ask questions, get things done, build reports and approve requests — all from one place, across every system you use. No more piecing it together by hand each morning.
Nothing without proof
Every fact shows exactly which system it came from and when. If that information is out of date, Verity blocks the action instead of using stale data.
Always asks first
A person approves the action first. Verity then carries it out and double-checks it actually worked before marking it done.
Measured, not claimed
268 / 268
unit tests passing, plus 45 / 45 store-level audit lines, independently verified
4,702 / 4,702
facts carrying a source and a confirmation time
0 of 108
rows stating a number with no evidence behind it
never
a blocked proposal reaching the underlying tool
Measured on the current build in our own sandbox, with simulated company data.
One ledger, two surfaces: the same decision ledger runs underneath Umpire and Verity.
Why now
RegulationA dated forcing function, not a trend.
FREE-AI framework
Seven sutras, six pillars, 26 recommendations for every RBI-regulated entity. Advisory today; its Governance and Assurance pillars ask for what a content filter cannot produce — a record for every decision, independent validation, and evidence when something goes wrong.
Regulation 16C
Binding today. Any SEBI-regulated entity is solely liable for the AI it uses — in-house or bought in — for data privacy and security, the integrity of AI output, and compliance with all applicable law. Liability has already moved to the buyer.
DPDP Rules 2025
Notified 13 Nov 2025 on an 18-month phased rollout: penalties and consent-manager registration from 13 Nov 2026, full substantive compliance by 13 May 2027, with penalties reaching ₹250 crore per violation. IRDAI guidance applies in parallel to insurers.
These are India’s forcing functions, shown here because they’re the sharpest example of what’s coming everywhere. The EU AI Act, GDPR, Singapore’s MAS and PDPA guidance, and sector regulators in most major markets are converging on the same asks: a record for every AI decision, independent validation, and evidence when something goes wrong.
Built for your industry — whichever it is
- Banking & financial institutions
- Insurance
- Capital markets
- Healthcare & life sciences
- Government & public sector
- Manufacturing
- Aerospace & defense
- Logistics & supply chain
- Retail & e-commerce
- Telecom
- Energy & utilities
- Real estate
- Tax & legal
Deployment and data posture
- On-premises, air-gapped if required
- Your own cloud tenancy, under your keys
- Hosted by us, in the country your rules require
- Regulated data stays inside your boundary at all times
- No client data trains anything outside your instance
- Complete network isolation on the regulated lane — proven, not promised
- Full audit logging; HIPAA controls, BAA on request
Let’s put this in front of your leadership team.
Proof before purchase: benchmarked on tests you design, on our own Harness, which you can rerun yourself. Pick one costly, repetitive workflow, run it in parallel with your team, and measure the delta. You keep the results either way.