Products — Vector & Value AI

Pramaan EUDR deforestation compliance for exporters: plot mapping, satellite checks, and buyer-ready due-diligence packages. Demo available
Token Bench See what a prompt costs across every major AI model before you send it, counted in your browser. Live
ReasonIQ Test whether an AI system’s reasoning holds up, not just whether its answer looks right. By conversation

Pramaan

EUDR deforestation compliance

Prove that every shipment to the EU came from land that wasn’t deforested after 2020, plot by plot, before your buyer asks.

30 December 2026: Large and medium operators must comply
30 June 2027: Micro and small enterprises must comply

The regulation. The EU Deforestation Regulation requires importers to prove that products weren’t grown on land deforested after 31 December 2020, using plot-level geolocation. It covers cattle, cocoa, coffee, palm oil, rubber, soy, and wood, and products made from them, such as leather, chocolate, tyres, and furniture. In May 2026 the European Commission confirmed it will not reopen the regulation, so the December 2026 date stands.

Who it’s for. Mid-sized exporters who sell into the EU and need to be ready without an enterprise compliance budget: coffee, rubber, leather, and wood-product exporters in India, and exporters of the same commodities across Southeast Asia, Africa, and Latin America.

What’s included

Supplier requests over WhatsApp, written in the supplier’s own language
Offline plot mapping for estates and plantations without mobile signal
Plot boundaries captured to EUDR rules: a polygon for any plot over 4 hectares
Satellite checks for forest loss after 31 December 2020 on every plot
Volume checks that flag shipments larger than the mapped plots can produce
Land records and legality documents collected and stored per plot
Due-diligence data and GeoJSON in the format the EU Information System accepts
A live share link so your EU buyer can review each shipment’s plots and checks

Priced for mid-sized exporters. Pay per plot mapped and per shipment prepared, with no annual enterprise contract.

Try the demo Book a readiness call

The demo uses sample data for a fictional coffee exporter and shows plot mapping, satellite checks, volume checks, supplier requests, and buyer packages. Regulation details: European Commission, EUDR implementation.

How Pramaan works

Map
Your suppliers get a WhatsApp request in their own language. Field agents map plot boundaries on a phone, even offline.
Check
Every plot is checked against satellite forest data for loss after 31 December 2020, and every shipment against what its plots can produce.
Package
Pramaan assembles the due-diligence data, land records, and GeoJSON in the format the EU Information System accepts.
Share
Your EU buyer gets a live link to each shipment’s plots and checks, instead of a folder of attachments.

What each commodity needs

Coffee

Many smallholder plots feed each lot. Every plot needs a location, and plots over 4 hectares need a full boundary.

Rubber

Covers natural rubber and products made from it, such as tyres. Most supply comes from small plantations that need mapping one by one.

Leather

Leather falls under cattle. Traceability runs back to the farms where the animals were kept, not only to the tannery.

Wood

Timber, furniture, and many wood handicrafts are covered. Each piece traces back to the plot where the wood was harvested.

Token Bench

LLM token and cost calculator

Paste a prompt and see its token count and price across every major model, before you send it.

The problem. AI costs are easy to underestimate. Every provider counts tokens differently and prices them differently, and the gap between a cheap model and an expensive one for the same prompt can be large.

Private by default. Counting runs in your browser with the providers’ published tokenizers. Your text isn’t uploaded unless you ask for an exact count from a provider that only offers one through its API.

What Token Bench does

Compares input cost across OpenAI, Anthropic, Google, Grok, DeepSeek, and open models
Totals tokens and cost for a whole batch of files or a code repository
Models the savings from prompt caching and trims filler from prompts
Estimates image, video, and audio tokens, with an API in free beta

Open Token Bench

ReasonIQ

Decision-intelligence platform

An AI reasoning and decision-intelligence platform designed to evaluate and improve complex reasoning capabilities.

The problem. Organizations increasingly rely on AI for decisions with real consequences: underwriting, forecasting, triage, resource allocation. Most evaluation methods check whether the final answer looks right. They don’t check whether the reasoning behind it would survive scrutiny from an expert, an auditor, or a regulator.

Who it’s for. Teams deploying AI into workflows where a plausible-sounding but wrong conclusion is expensive: risk, finance, operations, and product organizations building AI-assisted decision tools.

Why conventional evaluation falls short. Accuracy benchmarks reward the right answer for any reason. They miss reasoning that is fragile, shortcut-prone, or sensitive to how a problem happens to be phrased, which are exactly the failure modes that surface later, in production, on cases the benchmark never saw.

Key capabilities

Structured evaluation of reasoning chains, not just final outputs
Stress-testing against rephrasing, missing context, and adversarial framing
Traceable reasoning reports suitable for internal review or audit
Benchmarking across model versions as systems are upgraded

Ask about ReasonIQ

Example: health AI benchmark audits

Health AI companies increasingly compete on benchmark scores, and investors, hospitals, and journal reviewers rely on them. Small choices, such as which AI model grades the answers, which settings were on during the test, or whether the raw results were kept, can move a headline number by several points. ReasonIQ audits published health AI results at the level of code, data, and statistics, and reports which claims hold, which are fragile, and which can’t be verified.

Reproducibility: re-running a stratified sample end to end and comparing against the published scores
Grader sensitivity: re-grading identical answers with several AI graders to see how much the score depends on the choice
Grading variance: repeating the grading to separate real performance from grader randomness
Statistics: confidence intervals, paired comparisons, and correction for running many tests at once
Test conditions: whether benchmark runs used different settings, code paths, or safety features than the real product
Contamination: whether benchmark questions leaked into training data or the retrieval corpus
Robustness: whether scores survive reworded questions that mean the same thing
Provenance: whether every headline number traces back to item-level data that still exists

Scope
We agree which published claims are in scope and collect the code, data, results files, and exact run settings behind each one.
Reproduce
We re-run a stratified sample independently, with every setting logged, and compare the results item by item against what was published.
Stress-test
We re-grade with several graders, repeat grading, check for contamination and rewording effects, and re-derive the statistics.
Report
You receive a findings report rating each claim by confidence, with the evidence behind every rating and the fixes that would close each gap.

Ask about a health AI audit

Audits and publication support are offered as separate engagements, so an audit’s findings never depend on helping the same results get published.

Example: risk, validating AI in credit decisions

Lenders increasingly use AI to read financial statements and bank records, draft credit memos, and recommend whether to approve a loan. A recommendation can be right for the wrong reason: it may cite a ratio it calculated incorrectly, overlook a covenant breach, or change its answer when the same borrower is described differently. ReasonIQ tests the reasoning behind each recommendation, so credit committees, internal auditors, and regulators can see why the AI concluded what it did, and whether that reasoning holds.

Calculation checks: every ratio the AI cites, such as debt service coverage or leverage, recomputed from the source statements
Evidence tracing: each claim in the credit memo traced back to the document and page it came from
Consistency: the same borrower, presented in a different order, format, or wording, should get the same recommendation
Missing information: whether the AI flags gaps, like a missing year of accounts, or quietly fills them with assumptions
Fairness: whether the recommendation shifts when factors that shouldn’t matter, like a name or location, are changed
Policy adherence: every recommendation checked against the lender’s own credit policy and thresholds
Version drift: the same case files re-run whenever the underlying model is upgraded
Audit trail: a reasoning report for each decision, ready for credit committee or audit review

Ask about ReasonIQ for lending

Example: investment banking, checking AI-drafted valuation and deal materials

AI increasingly drafts pitchbooks, information memoranda, and valuation commentary. The risk is a polished narrative that doesn’t match the numbers, such as a summary that calls a company undervalued when the model says otherwise, or peer figures that appear in no filing. ReasonIQ checks whether each conclusion follows from the underlying analysis, traces every figure back to its source document, and tests whether conclusions shift sensibly when inputs such as the peer set or key assumptions change.

Ask about ReasonIQ for deal teams

Have a decision that warrants its own product?

We build deliberately, and announce new products when they’re ready for real use. If you have a high-value decision or workflow in mind, we’d like to hear about it.

Tell us about it

Have an AI problem worth solving?

Let’s identify where AI can create real economic and organizational value.

Start a conversation

A product line for decisions that carry real cost.