Agentic AI, data and the compute to run it.
Rillor builds AI agents and the datasets they depend on, applies them to financial and compute markets, and supplies, finances and deploys the GPU infrastructure behind them.
6
Datasets in the registry, each stating its sources, scope and refresh
Market, compute, legal and public-records data
Since 2018
U.S. equity and futures trade and quote history in our research warehouse, alongside DEX order-book data
Coverage window per dataset record
4
Index series: rental, availability, hardware and rent-to-value
Rillor Compute Index methodology v1.0
109
Access, admission and network tests passed for tenant isolation on our platform
Run 7 Oct 2026
What we do
Four lines of work that share one foundation: data we can trace back to its source.
AI
Agents that do real work.
Multi-agent AI systems that research, extract and evaluate, with each step checked against its source.
Data
Datasets with a paper trail.
Market data, GPU and neocloud pricing, legal and public records. Every record states its sources and scope.
Markets
Financial and compute markets.
Forecasting and simulation for financial markets, and price data and forward agreements for AI compute.
Compute
From specification to operation.
GPU systems, financing, data-center capacity and private AI clouds for AI teams.
Fields we work in
The disciplines behind our systems, datasets and compute work.
Agentic AI systems
Planners, specialist agents and audit steps that work together over large bodies of text and data.
AI evaluation and benchmarking
Black-box tests of LLM agents for bias, preferences and deception, including agents bidding in simulated markets and auctions.
Forecasting and analytics
Models that state a forecast before the outcome and are scored out of sample against baselines.
Document and regulation intelligence
Rulebooks, filings and contracts turned into structured, source-linked records and checkable models.
Data engineering
Collection, normalization and provenance for datasets whose sources change every day.
Financial market data
Order books, trades and quotes from decentralized exchanges, U.S. stocks and futures.
GPU infrastructure
Private AI clouds, model serving and the Kubernetes platforms they run on.
Compute pricing and finance
What GPU systems and cloud capacity cost, and how buyers finance them.
Six datasets, one standard
Each dataset states what it contains, where it comes from, how it stays current and what it can support.
| Dataset | What it holds | Used for | Link |
|---|---|---|---|
| Market data | Order-book snapshots and diffs, fills, liquidations and funding from on-chain perpetual futures venues. Trade and quote history for U.S. equities and futures. | Forecasting, simulation calibration, price-discovery research | Record |
| GPU cloud pricing and availability | Prices and availability that GPU cloud providers publish through their pricing APIs, by GPU class, region, term and pricing type. | Capacity planning, procurement, the Rillor Compute Index | Record |
| GPU systems pricing | System specifications, normalized configurations and price observations, kept apart by evidence class: listing, quote, reported sale, verified transaction. | Procurement comparison, valuation, financing diligence | Record |
| Public communications and market response | Public statements aligned in time with market moves, with random-baseline controls. | Event studies, news-driven simulation, methodology testing | Record |
| Legal datasets | Legal decisions and authorities with document structure and citations. | Legal research, citation analysis, retrieval systems | Record |
| Public procurement records | Contracting notices, awards and documents from public U.S. sources, with document text extraction and change tracking. | Supplier analysis, sourcing research, document AI | Record |
Need data that does not exist yet? We scope, build and maintain custom datasets.
Financial and compute markets
AI and data applied to how prices form, in financial markets and in the market for AI compute.
Rillor Compute Index
What AI compute costs, to rent and to own. The index tracks the prices and availability neoclouds and hyperscalers publish through their pricing APIs, and the prices of GPU systems themselves.
| Series | Measures | Unit |
|---|---|---|
| RCI Rental | Published rental prices by GPU class, region, term and pricing type | USD / GPU-hr |
| RCI Availability | Share of tracked offers providers report as available to rent | percent |
| RCI Hardware | GPU system prices by comparable grade | USD / system |
| RCI Rent-to-Value | Annualized rent over hardware price for the same GPU class | ratio |
Compute, from specification to operation
We help AI teams source, finance, house and run GPU infrastructure.
Systems we build
Our own products and platforms, built and operated by Rillor.
Market data warehouse
Trades and quotes for U.S. equities and futures, and order-book data from on-chain perpetual futures venues, aligned on one clock for research.
Used for calibrating simulated markets, forecasting research and price-discovery studies
Grellum
Agentic legal research. A planner assigns each question to specialist agents, an audit step reviews their work, and every citation must match retrieved evidence.
Every citation checked against the evidence it came from
The Rillor platform
Kubernetes, distributed storage, GitOps deployment and network-policy isolation. Every Rillor product runs on it.
109 access, admission and network tests passed for tenant isolation, 7 Oct 2026
MegaMCPs
Governed access to business tools for AI assistants, through an isolated MCP server for each customer.
One isolated server per customer, no shared tool runtime
How we work
The same rules apply to a dataset, a forecast and a GPU quote.
Every measured figure has a source and a date.
Records keep where they came from, when they were captured and what changed them.
Results are tested on data they were not built on.
Out-of-sample tests, baselines and random controls, with correction when many hypotheses are tested.
Every run can be replayed.
Model versions are pinned, inputs and outputs are logged, and important results are reimplemented independently.
Private deployments keep client data in the client's environment.
Models come to the data, on infrastructure the client controls.
Insights
Research notes, market reviews and full reports, each with its sources and a PDF of the full research.
Explainer
GPU financing structures explained
How leases, loans, sale-leasebacks, vendor and milestone financing, and contract-backed facilities pay for GPU systems, and what lenders check before they commit.
Market review
Compute market review: Q3 2026
Looking back at July to September 2026: buyers raised spending while short of capacity, newer GPUs rented for more, and compute futures went under CFTC review.
Research report
Cross-venue price discovery: regulated crude futures and on-chain perpetuals
Rillor tested whether CME WTI futures lead an on-chain crude perpetual. The lead held out of sample; no net edge after execution costs reached statistical significance.