Developer docs
API playgroundTry for free, no card

Search company profiles

Silicon Data

Full company profile

uuid00009h8

Namestring
Silicon Data
Legal namestring
Silicon Derivatives, Inc.
Company typeenum
Private
Founded yearint
2024
Descriptiontext

Silicon Data (legal entity Silicon Derivatives, Inc.) is a New York-headquartered GPU market intelligence platform founded in 2024. The company aggregates GPU pricing data across hyperscalers, neoclouds, brokers, and marketplaces, applying normalization across machine specs, interconnect, region, and contract terms to publish standardized daily indices, forward curves, performance benchmarks, carbon models, and token-economics datasets. Its core platform, SiliconNavigator™, provides tiered access (Market Pulse, Regional Pricing Feed, Instance-level Intelligence) covering 50 GPU chipsets with up to 8 years of historical data, while adjacent modules include SiliconIndex™ (with the SDH100RT published on Bloomberg), SiliconMark™ (vendor-neutral benchmarking), SiliconPriceIQ™ (AI-driven forecasting), SiliconCarbon™ (carbon footprint modeling), GPU Forward Curve (36-month no-arbitrage term structure), and Token Market Pulse / Token Pricebook covering 142+ LLM models. The company claims 3.5M+ aggregated data points, coverage of 95% of neocloud providers and 100% of major hyperscalers, and a client base of 500+ enterprise, financial, and semiconductor firms including several top-20 asset managers.

The platform architecture combines a proprietary data collection and normalization pipeline with financial-grade index methodology, validated against independent oversight and distributed natively through Bloomberg Terminal, Refinitiv (via dxFeed), Kaiko, and direct REST API access. The technology stack spans structured time-series data ingestion, code-based benchmark execution (SiliconMark), no-arbitrage curve construction, and AI-driven predictive analytics, with the GPU Forward Curve representing a methodological extension of sovereign yield and commodity curve construction applied to compute rentals.

Silicon Data monetizes through a hybrid model: a $499/month self-serve base portal subscription with module-based add-ons (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing — Raw Data); three-tier SiliconNavigator™ subscriptions scaling from entry-level Market Pulse to full instance-level intelligence for quant and allocator use cases; enterprise licensing of standardized indices via Bloomberg, Refinitiv, and direct API; and forward-dated compute futures economics anchored on its CME Group partnership. GTM is multi-motion: self-serve PLG with 7-day trial, enterprise field sales, API-first distribution, channel partnerships (Bloomberg, Refinitiv, dxFeed, Kaiko, CME), and event-driven moments around NVIDIA GTC and major index launches. The company is backed by $4.7M in seed funding co-led by DRW and Jump Trading Group (May 2025), and counts CME Group, Bloomberg, Refinitiv, Kaiko, NVIDIA, and TSMC among its strategic partners.

Short descriptiontext

Silicon Data is a GPU market intelligence platform that aggregates pricing, performance benchmarks, and forward curves across hyperscalers, neoclouds, and brokers, publishing standardized indices (e.g., SDH100RT on Bloomberg) for financial institutions, AI compute buyers, and data centers.

Operating statusenum
Operating
Ownership categoryenum
Headcount rangeband
11–50
akta.pro rankint
HeadquartersNew York, United States
HQ citystring
New York
HQ countrystring
United States
HQ regionstring
North America
Markets served

Serves global market

Offices3 records

Each record includes

City, Country, Type, Description, Source

Keyword5 values
GPU market intelligence, compute pricing data, GPU benchmarking, AI compute indices, token pricing analytics
Industry4 codes
1Networking & Data Center Silicon (Switch/NIC/DPU/SerDes)
CodeHDAHABABPrimaryNo
2Compute & Acceleration ICs (CPU/GPU/AI/FPGA/ASIC)
CodeHDAHABACPrimaryNo
3SAN Storage Arrays
CodeHDABAEAAPrimaryNo
4GPU/Accelerator Compute Servers
CodeHDABADACPrimaryNo
NAICS code3 codes
  • Computer Storage Device Manufacturing334112
  • Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services518210
  • Semiconductor and Other Electronic Component Manufacturing33441
SIC code2 codes
  • Semiconductors & Related Devices3674
  • Services-Computer Processing & Data Preparation7374
Product category
GPU Market Intelligence & Financial Index Data
Social media profiles2 records
GTM motion5 records

Each record includes

Type, Description, Source

Revenue model4 records
1Silicon Data Portal Subscription
TypeSubscription Recurring
Description

Recurring SaaS subscription at $499/month for base portal access with SiliconIndex, market direction tracking, PriceIQ predictive modeling, daily index values, NeoCloud vs Hyperscaler comparison, SiliconCarbon, and PDF performance reports; billed monthly with optional add-on modules (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing - Raw Data).

silicondata.com
2Tiered Data Subscriptions (SiliconNavigator™)
TypeSubscription Recurring
Description

Three-tier recurring subscriptions — Level 1 Market Pulse (free/entry), Level 2 Deep Pricing Feed (analyst/buyer), Level 3 Full Asset Intelligence (quant/allocator) — combining portal dashboard, API access, and CSV downloads of GPU pricing data; daily data volume scales from 50 to 2,000 average daily rows per tier.

silicondata.com
3Index Licensing & Data Distribution
TypeLicensing Royalties
Description

Licensing of GPU price indices (SDH100RT, A100, B200, Hyperscaler) and historical archives via Bloomberg Terminal, Refinitiv (through dxFeed), and direct API access for internal use or product creation by financial institutions and enterprises.

silicondata.com
4Raw Data Delivery (GPU Pricing - Raw Data)
TypeData Monetisation
Description

Flat file and API delivery of structured, machine-readable GPU and accelerator pricing datasets (~30,000 rows daily) with provider-level granularity and benchmark/market activity data for enterprise and fintech clients.

silicondata.com
Marketing channels8 records

Each record includes

Title, Type, Stage, Description, Source

Distribution channels7 records

Each record includes

Title, Type, Scope, Target buyer, Description, Source

Cost components5 values
Personnel, Technology or R&D, Infrastructure, Marketing or Sales, Operations
Pricing details5 tiers
1Silicon Data Portal Access — base subscription at $499/month with 7-day trial, includes SiliconIndex dashboard, daily index values, 90-day historical, Market direction tracking, PriceIQ predictive modeling, NeoCloud vs Hyperscaler comparison, SiliconCarbon, PDF performance reports.
ModelSubscriptionBilling cadenceMonthly
Notes

$499/month base; 7-day trial included; additional modules (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing - Raw Data) selectable as add-ons. Currently supports H100, H100-hyperscaler, A100, A100-hyperscaler, B200, MI300X, GDDR6 indices.

silicondata.com
2SiliconNavigator™ Level 1: Market Pulse — entry-tier with portal dashboard, global averages, retail vs rental, 7-day trendline, ~50 average daily data volume.
ModelSubscriptionBilling cadenceMonthly
Notes

Designed for generalists and enthusiasts; board reports and macro trend analysis; daily updates via portal dashboard.

silicondata.com
3SiliconNavigator™ Level 2: Deep Pricing Feed — portal + API + CSV download with platform-level and country-level pricing, one year historical, ~300 average daily rows.
ModelSubscriptionBilling cadenceMonthly
Notes

Aimed at analysts, buyers, and strategists; covers spot (on-demand), interruptible, 3-/6-month reserved, second-hand, and refurbished GPUs across U.S. and other regions.

silicondata.com
4SiliconNavigator™ Level 3: Full Asset Intelligence — instance-level raw data, hyper-local datacenter-level pricing, up to 8 years historical, ~2,000 average daily rows.
ModelSubscriptionBilling cadenceMonthly
Notes

Targeted at quant funds, infrastructure leaders, and allocators; includes raw asset visibility with detailed specs, machine IDs, geographies, pricing, and usage data.

silicondata.com
5Add-on modules: GPU Index Feed, GPU Forward Curve, GPU Pricing by Country, GPU Pricing - Raw Data, Token Market Pulse, Token Pricebook, Market Trends.
ModelHybridBilling cadenceMonthly
Notes

Configurable add-ons atop base subscription; enterprise plans via Talk to Sales for licensing and custom deployments.

silicondata.com
GTM typeB2B
B2B
Offering typeSoftware
Software
Brand1 of 6 records shown
1SiliconNavigator™
Description

Real-time GPU pricing and market intelligence platform providing tiered access to GPU rental and sales pricing data across clouds, brokers, and GPU design houses.

silicondata.com
+5 more records
Core offering1 text field

Silicon Data operates a GPU compute market intelligence platform that aggregates 3.5M+ pricing data points across hyperscalers, neo-clouds, brokers, and marketplaces. It publishes standardized daily GPU price indices (SDH100RT, SDA100RT, SDB200RT), performance benchmarks, forward curves, token pricing datasets, and carbon footprint models, delivered via a self-serve web portal, REST API, and distribution through Bloomberg Terminal, Refinitiv, and Kaiko.

Differentiator
Functional benefit
Problem solved
Product overview1 text field

Silicon Data operates a unified GPU compute intelligence platform-plus-modules architecture, anchored by the core SiliconNavigator™ market intelligence platform and extended through a portfolio of named sub-products: SiliconMark™ (benchmarking), SiliconIndex™ (standardized pricing indices such as SDH100RT and SDB200RT), SiliconPriceIQ™ (AI-powered pricing optimization and forecasts), SiliconCarbon™ (carbon footprint calculation), GPU Forward Curve (term-structure forward pricing), RAM Index, Token Market Pulse, and Token Pricebook. Customers access the offering via Silicon Data Portal Access (base package, $499/month) and unlock additional modules (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing – Raw Data) on top. Distribution of standardized indices extends through Bloomberg, Refinitiv, and Kaiko, with dxFeed providing market-data infrastructure and index management integration; Silicon Data's GPU indices also underpin CME Group's first compute futures contracts.

Product and service11 records
1SiliconNavigator™
CategoryCore platform — GPU market intelligence
Description

Core GPU pricing and market intelligence platform that aggregates 100+ data sources across cloud, marketplace, and bare-metal channels to surface GPU compute pricing, availability, configuration, and historical trends for compute buyers, sellers, and investors. Provides tiered access (Level 1 Market Pulse, Level 2 Deep Pricing Feed, Level 3 Full Asset Intelligence) with up to 8 years of historical data and API/portal delivery.

2SiliconMark™
CategoryAdd-on / module — GPU performance benchmarking
Description

Vendor-neutral GPU benchmarking solution that executes standardized workloads (FP32, FP16, memory bandwidth, GPU/cluster/LLM performance) to measure real-world performance, detect drift and variability over time, and produce benchmark certificates and a public scoreboard for cross-provider comparison. Supports pricing analytics, SLAs, and procurement validation.

3SiliconIndex™ (Silicon Data Indices: SDH100RT, SDA100RT, SDB200RT, Hyperscaler Indices)
CategoryAdd-on / module — standardized price indices
Description

Standardized, exchange-grade GPU price indexing service delivering chip-agnostic indices including the H100 Rental Index (SDH100RT, on Bloomberg), A100 Index (SDA100RT), B200 Index (SDB200RT), and A100/H100 Hyperscaler Rental Indices. Built with disclosed normalization methodology, outlier removal, and independent validation for institutional financial use.

4SiliconPriceIQ™
CategoryAdd-on / module — AI pricing forecasting
Description

AI-powered GPU pricing optimization and predictive analytics engine that forecasts rental market trends and prices across 100+ data sources, predicting pricing by model, configuration, region, and rental term (on-demand, reserved, spot). Recommends optimal rental, reservation, and purchasing decisions for compute buyers and sellers, with daily updates and REST API access.

5SiliconCarbon™
CategoryAdd-on / module — carbon footprint analytics
Description

Real-time carbon footprint calculator for GPU-intensive compute workflows that quantifies emissions (kgCO2e) per job/chip/hour/region using regional grid intensity, GPU TDP/power draw, and workload inputs. Supports carbon-aware workload placement, sustainability reporting, and ESG mandates for financed AI infrastructure portfolios. Delivered via web tool and enterprise API.

6GPU Forward Curve
CategoryAdd-on / module — term-structure forward pricing
Description

Term-structure pricing product that derives implied forward rates for GPU rentals using no-arbitrage methodology across monthly and quarterly tenors over a 36-month horizon. Supports swap pricing, covenant modeling, parametric insurance triggers, hedging, contract structuring, and risk management for hyperscalers, neoclouds, and financial institutions.

7Silicon Data Portal Access
CategoryBase access package — portal subscription
Description

Base subscription package priced at $499/month with a 7-day trial, providing entry to the SiliconNavigator dashboard, daily index values, 90-day historical, market direction tracking, PriceIQ predictive modeling, NeoCloud vs Hyperscaler comparison, SiliconCarbon, and PDF performance reports. Add-on modules (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing - Raw Data) are available for additional fees.

8RAM Index
CategoryAdd-on / module — memory pricing index
Description

Tracks GDDR6 price-per-GB as a complement to GPU pricing intelligence, providing visibility into GPU memory cost trends.

9Token Market Pulse
CategoryAdd-on / module — LLM token pricing
Description

Daily-updated view of model token pricing across major providers, tracking 142+ LLM models with standardized pricing, cost/adoption/momentum panels, lifecycle coverage (active/legacy/deprecated), and price-performance and market-share views for AI model economics.

10Token Pricebook
CategoryAdd-on / module — LLM token pricing benchmark
Description

Unified, financial-grade standard for AI model economics covering 142+ models, enabling consistent comparison and benchmarking of token pricing across providers and modalities.

11GPU Pricing - Raw Data
CategoryAdd-on / module — raw data feed
Description

Flat file and API delivery of structured, machine-readable GPU and accelerator pricing datasets with approximately 30,000 rows daily, provider-level granularity, and benchmark/market activity data for enterprise and fintech clients.

Scale indicator14 records

Each record includes

Type, Value, Description, Source

Partnership11 partners
Strategic tierCoreTypeGTM or Marketing PartnerAnnounced on2026-06-16
Description

Within days of CME Group and Silicon Data's compute futures announcement, ProShares filed a proposal with the SEC for an exchange-traded fund tied to the proposed compute futures contracts, signaling early investor demand for AI compute as a tradeable asset class.

Strategic tierCoreTypeGTM or Marketing PartnerAnnounced on2026-06-16
Description

Within days of CME Group and Silicon Data's compute futures announcement, Rex Shares filed a proposal with the SEC for an exchange-traded fund tied to the proposed compute futures contracts, signaling early investor demand for AI compute as a tradeable asset class.

Strategic tierFlagshipTypeStrategic or Co-development PartnerAnnounced on2026-05-12
Description

Jointly announced the launch of the world's first compute futures market on May 12, 2026 (Chicago). CME will list cash-settled derivatives contracts tied to Silicon Data's daily GPU benchmark indices for on-demand rental rates across A100, H100, and B200. Pending CFTC regulatory review, launch targeted for H2 2026. The contracts will allow traders, financial institutions, AI builders, and cloud providers to hedge volatility and price risk in the multi-trillion-dollar compute market.

Strategic tierMinorTypeStrategic or Co-development PartnerAnnounced on2026-04-23
Description

Joint research with the College of William & Mary and Jefferson Lab published via IEEE Spectrum revealed significant variability in performance among GPUs of the same model, with the study testing 3,500 GPUs across multiple providers and finding performance differences up to 38% due to manufacturing variations.

Strategic tierCoreTypeStrategic or Co-development PartnerAnnounced on2026-04-20
Description

Silicon Data cites a partnership with NVIDIA providing upstream visibility into the supply side of the GPU forward curve. NVIDIA's chip generations and pricing decisions (e.g., H100, B200, B200 MSRP revisions in Feb 2026) directly influence Silicon Data's index methodology.

Strategic tierCoreTypeStrategic or Co-development PartnerAnnounced on2026-04-20
Description

Silicon Data cites a partnership with TSMC providing upstream visibility into the supply side of the GPU forward curve. TSMC's manufacturing capacity and packaging constraints feed into Silicon Data's supply-side modeling for GPU rental price forecasting.

Strategic tierFlagshipTypeStrategic or Co-development PartnerAnnounced on2025-10-22
Description

Compute Exchange, an AI compute infrastructure marketplace led by Silicon Data CEO Carmen Li, operates alongside Silicon Data as a sister company. DRW is also a founding backer of Compute Exchange. The two companies collaborate on GPU pricing index (Silicon Data) and a spot marketplace for GPU procurement (Compute Exchange), with Compute Exchange launching its RFQ Hub and Partnership & Referral Program in October 2025.

Strategic tierCoreTypeTechnology or IntegrationAnnounced on2025-10-08
Description

Market data and index management services provider dxFeed (named Best Data Provider 2025 by Fund Intelligence Operations and Services Awards) integrated Silicon Data's H100 and A100 Indices into Refinitiv terminals; dxFeed supports distribution, API integration, and daily index recalculation for Silicon Data's indices.

Strategic tierFlagshipTypeChannel Partner/ Reseller/ DistributorAnnounced on2025-07-31
Description

By end of July 2025, both the flagship H100 Index and newly launched A100 Index became accessible to Refinitiv users, with dxFeed providing market data and index management integration. Designed to deliver transparency into AI compute economics for institutional investors, infrastructure consumers, and policymakers.

Strategic tierFlagshipTypeChannel Partner/ Reseller/ DistributorAnnounced on2025-05-20
Description

Silicon Data publishes its GPU indices on the Bloomberg Terminal, with the H100 Rental Index listed under ticker SDH100RT. Indices are built to financial-index standards and serve institutional users including asset managers, exchanges, and hedge funds.

Strategic tierCoreTypeChannel Partner/ Reseller/ DistributorAnnounced on2025-05-20
Description

Silicon Data is working with Kaiko, a leading market data provider, to distribute the H100 Rental Index and additional indices to a broader set of market participants beyond Bloomberg and Refinitiv.

Recent move6 records

Each record includes

Date, Type, Title, Description, Source

Expansion highlight6 records

Each record includes

Type, Description

Peers10 records
TypeDirect peer
Description

Sister company also led by CEO Carmen Li and backed by DRW; operates a spot marketplace for GPU procurement. Directly comparable because it sits at the intersection of GPU pricing transparency and physical GPU compute procurement, sharing data flows and customers with Silicon Data.

TypeDirect peer
Description

Digital asset market data provider distributing institutional-grade indices on financial terminals. Directly comparable because Kaiko is itself a distribution partner for Silicon Data's H100 Rental Index and operates the same business model of normalizing fragmented venue data into standardized indices for financial institutions.

TypeDirect peer
Description

Cloud cost optimization and FinOps platform that automates GPU and compute spending across hyperscalers. Directly comparable because it serves the same compute-buyer persona with pricing intelligence and procurement optimization, competing for the same enterprise budget line.

TypeDirect peer
Description

Cloud cost intelligence platform for engineering and finance teams managing AWS/Azure/GCP spend. Comparable because it targets the same FinOps buyer with cloud-pricing analytics and budget forecasting, and competes for procurement-tooling wallet share.

TypeDirect peer
Description

Industry-standard GPU and AI benchmarking consortium maintained by MLCommons. Comparable because SiliconMark operates in the same GPU benchmarking category with comparable FP32/FP16/memory bandwidth tests and published scoreboards, targeting overlapping audiences of cloud buyers and hardware providers.

TypeBroad incumbent
Description

Large-scale GPU cloud provider offering H100/B200 instances and managed AI infrastructure. Comparable as a primary data source within SiliconNavigator's coverage and as a customer for benchmarking; overlapping as both a neocloud the platform tracks and a potential partner.

TypeBroad incumbent
Description

GPU neocloud built on stranded/clean energy, offering large-scale H100/B200 rentals. Comparable as a major neocloud in SiliconNavigator's covered universe and as a likely buyer of benchmarking/forecasting tools for its own sales motion.

TypeBroad incumbent
Description

GPU cloud provider specializing in AI training and inference on H100/B200 hardware. Comparable as a covered neocloud provider and as a customer with parallel benchmarking/procurement-intelligence needs to those Silicon Data addresses.

TypeBroad incumbent
Description

AI cloud platform providing GPU-backed inference and training infrastructure. Comparable as a covered neocloud and as a participant in the same token-pricing ecosystem tracked by Token Market Pulse and Token Pricebook.

TypeEmerging player
Description

GPU neocloud offering large-scale H100/B200 capacity with focus on AI labs. Comparable as an emerging neocloud in SiliconNavigator's covered universe and as a buyer of market intelligence to inform its own pricing.

Market position
Strengths5 records

Each record includes

Headline, Details, Source

Weaknesses2 records

Each record includes

Headline, Details, Source

Competitive moat7 records

Each record includes

Type, Details

Key risks7 records

Each record includes

Headline, Details, Source

Key highlights7 records

Each record includes

Headline, Details, Source

Customer concentration

Classification, Details

Segment5 records

Each record includes

Title, Type, Primary, Description, Pain point addressed, Use case, Source

Ideal customer profile4 records

Each record includes

Profile, Firmographic size, Sales motion, Sales cycle length, Buying structure, Purchase trigger, Buyer persona, Geography, Industry vertical, Primary use case, Description, Pain points, Evidence proof points, Target buyer

Technology focused
Yes
API detail
Has APIbool
Yes

Docs URL, Description

Integration4 records

Each record includes

Title, Type, Description, Source

AI capability6 records

Each record includes

Type, Description, Source

AI maturity
App detail

Has app

Feature7 records

Each record includes

Title, Differentiator, Description, Source

Core technology
Revenue estimate
Valuation estimate
Number of profiles
Profiles2 records

Each record includes

Name, Designation, Designation category, Overview, Profile commentary, Source

Subsidiaries1 record

Each record includes

Name, Acquired on, Relationship type, Type, Business focus

Compliance2 records

Each record includes

Name, Class, Description

Funding overview

Funding stage, Last funding date, Total funding USD

Funding rounds2 records

Each record includes

Round, Amount USD, Date, Pre money valuation, Total investors, Investors, News

Investors18 records

Each record includes

Name, Type, Date of entry, Rounds participated, Website

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Each record includes

Name, Acquisition type, Announced date, Completed date, Status, Website, News

Investment

Each record includes

Name, Round, Announced date, Lead investor, Website, News

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Silicon Data

GPU Market Intelligence & Financial Index Datasilicondata.com

Silicon Data is a GPU market intelligence platform that aggregates pricing, performance benchmarks, and forward curves across hyperscalers, neoclouds, and brokers, publishing standardized indices (e.g., SDH100RT on Bloomberg) for financial institutions, AI compute buyers, and data centers.

What Silicon Data does

Silicon Data (legal entity Silicon Derivatives, Inc.) is a New York-headquartered GPU market intelligence platform founded in 2024. The company aggregates GPU pricing data across hyperscalers, neoclouds, brokers, and marketplaces, applying normalization across machine specs, interconnect, region, and contract terms to publish standardized daily indices, forward curves, performance benchmarks, carbon models, and token-economics datasets. Its core platform, SiliconNavigator™, provides tiered access (Market Pulse, Regional Pricing Feed, Instance-level Intelligence) covering 50 GPU chipsets with up to 8 years of historical data, while adjacent modules include SiliconIndex™ (with the SDH100RT published on Bloomberg), SiliconMark™ (vendor-neutral benchmarking), SiliconPriceIQ™ (AI-driven forecasting), SiliconCarbon™ (carbon footprint modeling), GPU Forward Curve (36-month no-arbitrage term structure), and Token Market Pulse / Token Pricebook covering 142+ LLM models. The company claims 3.5M+ aggregated data points, coverage of 95% of neocloud providers and 100% of major hyperscalers, and a client base of 500+ enterprise, financial, and semiconductor firms including several top-20 asset managers.

The platform architecture combines a proprietary data collection and normalization pipeline with financial-grade index methodology, validated against independent oversight and distributed natively through Bloomberg Terminal, Refinitiv (via dxFeed), Kaiko, and direct REST API access. The technology stack spans structured time-series data ingestion, code-based benchmark execution (SiliconMark), no-arbitrage curve construction, and AI-driven predictive analytics, with the GPU Forward Curve representing a methodological extension of sovereign yield and commodity curve construction applied to compute rentals.

Silicon Data monetizes through a hybrid model: a $499/month self-serve base portal subscription with module-based add-ons (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing — Raw Data); three-tier SiliconNavigator™ subscriptions scaling from entry-level Market Pulse to full instance-level intelligence for quant and allocator use cases; enterprise licensing of standardized indices via Bloomberg, Refinitiv, and direct API; and forward-dated compute futures economics anchored on its CME Group partnership. GTM is multi-motion: self-serve PLG with 7-day trial, enterprise field sales, API-first distribution, channel partnerships (Bloomberg, Refinitiv, dxFeed, Kaiko, CME), and event-driven moments around NVIDIA GTC and major index launches. The company is backed by $4.7M in seed funding co-led by DRW and Jump Trading Group (May 2025), and counts CME Group, Bloomberg, Refinitiv, Kaiko, NVIDIA, and TSMC among its strategic partners.

Silicon Data firmographics

Firmographics
Name
Silicon Data
Legal name
Silicon Derivatives, Inc.
Website
https://www.silicondata.com
Company type
Private
Founded year
2024
Operating status
Operating
Headcount range
11–50 employees
Short description
Silicon Data is a GPU market intelligence platform that aggregates pricing, performance benchmarks, and forward curves across hyperscalers, neoclouds, and brokers, publishing standardized indices (e.g., SDH100RT on Bloomberg) for financial institutions, AI compute buyers, and data centers.
Ownership category
akta.pro rank

Where Silicon Data is headquartered

Location

Headquarters

HQ city
New York
HQ country
United States
HQ region
North America

Offices3 records

Markets served

Silicon Data business model

Business model
GTM type
B2B
Offering type
Software
Cost components
Personnel, Technology or R&D, Infrastructure, Marketing or Sales, Operations

Revenue model

  1. Silicon Data Portal Subscription: Recurring SaaS subscription at $499/month for base portal access with SiliconIndex, market direction tracking, PriceIQ predictive modeling, daily index values, NeoCloud vs Hyperscaler comparison, SiliconCarbon, and PDF performance reports; billed monthly with optional add-on modules (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing - Raw Data).
  2. Tiered Data Subscriptions (SiliconNavigator™): Three-tier recurring subscriptions — Level 1 Market Pulse (free/entry), Level 2 Deep Pricing Feed (analyst/buyer), Level 3 Full Asset Intelligence (quant/allocator) — combining portal dashboard, API access, and CSV downloads of GPU pricing data; daily data volume scales from 50 to 2,000 average daily rows per tier.
  3. Index Licensing & Data Distribution: Licensing of GPU price indices (SDH100RT, A100, B200, Hyperscaler) and historical archives via Bloomberg Terminal, Refinitiv (through dxFeed), and direct API access for internal use or product creation by financial institutions and enterprises.
  4. Raw Data Delivery (GPU Pricing - Raw Data): Flat file and API delivery of structured, machine-readable GPU and accelerator pricing datasets (~30,000 rows daily) with provider-level granularity and benchmark/market activity data for enterprise and fintech clients.

Pricing tiers

ModelBillingPrice
SubscriptionMonthlySilicon Data Portal Access — base subscription at $499/month with 7-day trial, includes SiliconIndex dashboard, daily index values, 90-day historical, Market direction tracking, PriceIQ predictive modeling, NeoCloud vs Hyperscaler comparison, SiliconCarbon, PDF performance reports.
SubscriptionMonthlySiliconNavigator™ Level 1: Market Pulse — entry-tier with portal dashboard, global averages, retail vs rental, 7-day trendline, ~50 average daily data volume.
SubscriptionMonthlySiliconNavigator™ Level 2: Deep Pricing Feed — portal + API + CSV download with platform-level and country-level pricing, one year historical, ~300 average daily rows.
SubscriptionMonthlySiliconNavigator™ Level 3: Full Asset Intelligence — instance-level raw data, hyper-local datacenter-level pricing, up to 8 years historical, ~2,000 average daily rows.
HybridMonthlyAdd-on modules: GPU Index Feed, GPU Forward Curve, GPU Pricing by Country, GPU Pricing - Raw Data, Token Market Pulse, Token Pricebook, Market Trends.

Go-to-market motion5 records

Distribution channels7 records

Marketing channels8 records

Silicon Data product offering

Product offering

Core offering

Silicon Data operates a GPU compute market intelligence platform that aggregates 3.5M+ pricing data points across hyperscalers, neo-clouds, brokers, and marketplaces. It publishes standardized daily GPU price indices (SDH100RT, SDA100RT, SDB200RT), performance benchmarks, forward curves, token pricing datasets, and carbon footprint models, delivered via a self-serve web portal, REST API, and distribution through Bloomberg Terminal, Refinitiv, and Kaiko.

Product overview

Silicon Data operates a unified GPU compute intelligence platform-plus-modules architecture, anchored by the core SiliconNavigator™ market intelligence platform and extended through a portfolio of named sub-products: SiliconMark™ (benchmarking), SiliconIndex™ (standardized pricing indices such as SDH100RT and SDB200RT), SiliconPriceIQ™ (AI-powered pricing optimization and forecasts), SiliconCarbon™ (carbon footprint calculation), GPU Forward Curve (term-structure forward pricing), RAM Index, Token Market Pulse, and Token Pricebook. Customers access the offering via Silicon Data Portal Access (base package, $499/month) and unlock additional modules (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing – Raw Data) on top. Distribution of standardized indices extends through Bloomberg, Refinitiv, and Kaiko, with dxFeed providing market-data infrastructure and index management integration; Silicon Data's GPU indices also underpin CME Group's first compute futures contracts.

Differentiator

Problem solved

Functional benefit

Brands

  • SiliconNavigator™: Real-time GPU pricing and market intelligence platform providing tiered access to GPU rental and sales pricing data across clouds, brokers, and GPU design houses.
  • SiliconMark™
  • SiliconIndex™
  • SiliconPriceIQ™
  • SiliconCarbon™
  • GPU Forward Curve

Products and services

  • SiliconNavigator™ Core GPU pricing and market intelligence platform that aggregates 100+ data sources across cloud, marketplace, and bare-metal channels to surface GPU compute pricing, availability, configuration, and historical trends for compute buyers, sellers, and investors. Provides tiered access (Level 1 Market Pulse, Level 2 Deep Pricing Feed, Level 3 Full Asset Intelligence) with up to 8 years of historical data and API/portal delivery.
  • SiliconMark™ Vendor-neutral GPU benchmarking solution that executes standardized workloads (FP32, FP16, memory bandwidth, GPU/cluster/LLM performance) to measure real-world performance, detect drift and variability over time, and produce benchmark certificates and a public scoreboard for cross-provider comparison. Supports pricing analytics, SLAs, and procurement validation.
  • SiliconIndex™ (Silicon Data Indices: SDH100RT, SDA100RT, SDB200RT, Hyperscaler Indices) Standardized, exchange-grade GPU price indexing service delivering chip-agnostic indices including the H100 Rental Index (SDH100RT, on Bloomberg), A100 Index (SDA100RT), B200 Index (SDB200RT), and A100/H100 Hyperscaler Rental Indices. Built with disclosed normalization methodology, outlier removal, and independent validation for institutional financial use.
  • SiliconPriceIQ™ AI-powered GPU pricing optimization and predictive analytics engine that forecasts rental market trends and prices across 100+ data sources, predicting pricing by model, configuration, region, and rental term (on-demand, reserved, spot). Recommends optimal rental, reservation, and purchasing decisions for compute buyers and sellers, with daily updates and REST API access.
  • SiliconCarbon™ Real-time carbon footprint calculator for GPU-intensive compute workflows that quantifies emissions (kgCO2e) per job/chip/hour/region using regional grid intensity, GPU TDP/power draw, and workload inputs. Supports carbon-aware workload placement, sustainability reporting, and ESG mandates for financed AI infrastructure portfolios. Delivered via web tool and enterprise API.
  • GPU Forward Curve Term-structure pricing product that derives implied forward rates for GPU rentals using no-arbitrage methodology across monthly and quarterly tenors over a 36-month horizon. Supports swap pricing, covenant modeling, parametric insurance triggers, hedging, contract structuring, and risk management for hyperscalers, neoclouds, and financial institutions.
  • Silicon Data Portal Access Base subscription package priced at $499/month with a 7-day trial, providing entry to the SiliconNavigator dashboard, daily index values, 90-day historical, market direction tracking, PriceIQ predictive modeling, NeoCloud vs Hyperscaler comparison, SiliconCarbon, and PDF performance reports. Add-on modules (Market Trends, GPU Forward Curve, Token Market Pulse, Token Pricebook, GPU Index Feed, GPU Pricing by Country, GPU Pricing - Raw Data) are available for additional fees.
  • RAM Index Tracks GDDR6 price-per-GB as a complement to GPU pricing intelligence, providing visibility into GPU memory cost trends.
  • Token Market Pulse Daily-updated view of model token pricing across major providers, tracking 142+ LLM models with standardized pricing, cost/adoption/momentum panels, lifecycle coverage (active/legacy/deprecated), and price-performance and market-share views for AI model economics.
  • Token Pricebook Unified, financial-grade standard for AI model economics covering 142+ models, enabling consistent comparison and benchmarking of token pricing across providers and modalities.
  • GPU Pricing - Raw Data Flat file and API delivery of structured, machine-readable GPU and accelerator pricing datasets with approximately 30,000 rows daily, provider-level granularity, and benchmark/market activity data for enterprise and fintech clients.

Companies that use Silicon Data

Customer profile

Segments5 records

Ideal customer profiles4 records

Silicon Data technology and API

Technology

Technology focussed Yes

API detail

Has API
Yes
API docs
API detail

Core technology

AI maturity

App detail

Integration4 records

AI capability6 records

Feature7 records

Silicon Data partnerships and signals

Strategic signal

Partnerships

Eleven partnerships are on record, tiered core, flagship and minor.

  • ProSharescoreGTM or Marketing Partner · 16 June 2026Within days of CME Group and Silicon Data's compute futures announcement, ProShares filed a proposal with the SEC for an exchange-traded fund tied to the proposed compute futures contracts, signaling early investor demand for AI compute as a tradeable asset class.
  • Rex SharescoreGTM or Marketing Partner · 16 June 2026Within days of CME Group and Silicon Data's compute futures announcement, Rex Shares filed a proposal with the SEC for an exchange-traded fund tied to the proposed compute futures contracts, signaling early investor demand for AI compute as a tradeable asset class.
  • CME GroupflagshipStrategic or Co-development Partner · 12 May 2026Jointly announced the launch of the world's first compute futures market on May 12, 2026 (Chicago). CME will list cash-settled derivatives contracts tied to Silicon Data's daily GPU benchmark indices for on-demand rental rates across A100, H100, and B200. Pending CFTC regulatory review, launch targeted for H2 2026. The contracts will allow traders, financial institutions, AI builders, and cloud providers to hedge volatility and price risk in the multi-trillion-dollar compute market.
  • College of William & Mary and Jefferson LabminorStrategic or Co-development Partner · 23 April 2026Joint research with the College of William & Mary and Jefferson Lab published via IEEE Spectrum revealed significant variability in performance among GPUs of the same model, with the study testing 3,500 GPUs across multiple providers and finding performance differences up to 38% due to manufacturing variations.
  • NVIDIAcoreStrategic or Co-development Partner · 20 April 2026Silicon Data cites a partnership with NVIDIA providing upstream visibility into the supply side of the GPU forward curve. NVIDIA's chip generations and pricing decisions (e.g., H100, B200, B200 MSRP revisions in Feb 2026) directly influence Silicon Data's index methodology.
  • TSMCcoreStrategic or Co-development Partner · 20 April 2026Silicon Data cites a partnership with TSMC providing upstream visibility into the supply side of the GPU forward curve. TSMC's manufacturing capacity and packaging constraints feed into Silicon Data's supply-side modeling for GPU rental price forecasting.
  • Compute ExchangeflagshipStrategic or Co-development Partner · 22 October 2025Compute Exchange, an AI compute infrastructure marketplace led by Silicon Data CEO Carmen Li, operates alongside Silicon Data as a sister company. DRW is also a founding backer of Compute Exchange. The two companies collaborate on GPU pricing index (Silicon Data) and a spot marketplace for GPU procurement (Compute Exchange), with Compute Exchange launching its RFQ Hub and Partnership & Referral Program in October 2025.
  • dxFeedcoreTechnology or Integration · 8 October 2025Market data and index management services provider dxFeed (named Best Data Provider 2025 by Fund Intelligence Operations and Services Awards) integrated Silicon Data's H100 and A100 Indices into Refinitiv terminals; dxFeed supports distribution, API integration, and daily index recalculation for Silicon Data's indices.
  • RefinitivflagshipChannel Partner/ Reseller/ Distributor · 31 July 2025By end of July 2025, both the flagship H100 Index and newly launched A100 Index became accessible to Refinitiv users, with dxFeed providing market data and index management integration. Designed to deliver transparency into AI compute economics for institutional investors, infrastructure consumers, and policymakers.
  • BloombergflagshipChannel Partner/ Reseller/ Distributor · 20 May 2025Silicon Data publishes its GPU indices on the Bloomberg Terminal, with the H100 Rental Index listed under ticker SDH100RT. Indices are built to financial-index standards and serve institutional users including asset managers, exchanges, and hedge funds.
  • KaikocoreChannel Partner/ Reseller/ Distributor · 20 May 2025Silicon Data is working with Kaiko, a leading market data provider, to distribute the H100 Rental Index and additional indices to a broader set of market participants beyond Bloomberg and Refinitiv.

Scale indicators14 records

Recent moves6 records

Expansion highlights6 records

Silicon Data competitors and assessment

Company assessment

Direct peers

  • Compute Exchange: Sister company also led by CEO Carmen Li and backed by DRW; operates a spot marketplace for GPU procurement. Directly comparable because it sits at the intersection of GPU pricing transparency and physical GPU compute procurement, sharing data flows and customers with Silicon Data.
  • Kaiko: Digital asset market data provider distributing institutional-grade indices on financial terminals. Directly comparable because Kaiko is itself a distribution partner for Silicon Data's H100 Rental Index and operates the same business model of normalizing fragmented venue data into standardized indices for financial institutions.
  • Cast.ai: Cloud cost optimization and FinOps platform that automates GPU and compute spending across hyperscalers. Directly comparable because it serves the same compute-buyer persona with pricing intelligence and procurement optimization, competing for the same enterprise budget line.
  • CloudZero: Cloud cost intelligence platform for engineering and finance teams managing AWS/Azure/GCP spend. Comparable because it targets the same FinOps buyer with cloud-pricing analytics and budget forecasting, and competes for procurement-tooling wallet share.
  • MLPerf: Industry-standard GPU and AI benchmarking consortium maintained by MLCommons. Comparable because SiliconMark operates in the same GPU benchmarking category with comparable FP32/FP16/memory bandwidth tests and published scoreboards, targeting overlapping audiences of cloud buyers and hardware providers.

Broad incumbents

  • CoreWeave: Large-scale GPU cloud provider offering H100/B200 instances and managed AI infrastructure. Comparable as a primary data source within SiliconNavigator's coverage and as a customer for benchmarking; overlapping as both a neocloud the platform tracks and a potential partner.
  • Crusoe Energy: GPU neocloud built on stranded/clean energy, offering large-scale H100/B200 rentals. Comparable as a major neocloud in SiliconNavigator's covered universe and as a likely buyer of benchmarking/forecasting tools for its own sales motion.
  • Lambda Labs: GPU cloud provider specializing in AI training and inference on H100/B200 hardware. Comparable as a covered neocloud provider and as a customer with parallel benchmarking/procurement-intelligence needs to those Silicon Data addresses.
  • Together AI: AI cloud platform providing GPU-backed inference and training infrastructure. Comparable as a covered neocloud and as a participant in the same token-pricing ecosystem tracked by Token Market Pulse and Token Pricebook.

Emerging players

  • Fluidstack: GPU neocloud offering large-scale H100/B200 capacity with focus on AI labs. Comparable as an emerging neocloud in SiliconNavigator's covered universe and as a buyer of market intelligence to inform its own pricing.

Market position

Strengths5 records

Weaknesses2 records

Competitive moat7 records

Key risks7 records

Key highlights7 records

Customer concentration

Silicon Data social profiles

Digital presence

Silicon Data compliance and trust

Trust signal

Compliance2 records

Silicon Data financial estimates

Financial estimate

Revenue estimate

Valuation estimate

Silicon Data leadership team

Management profile

Number of profiles

Profiles2 records

Silicon Data subsidiaries and ownership

Company hierarchy

Subsidiaries1 record

Silicon Data funding detail

Funding detail

Funding overview

Funding rounds2 records

Investors18 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

Silicon Data M&A and investment

M&A and investment

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Silicon Data

What does Silicon Data do?

Silicon Data operates a GPU compute market intelligence platform that aggregates 3.5M+ pricing data points across hyperscalers, neo-clouds, brokers, and marketplaces. It publishes standardized daily GPU price indices (SDH100RT, SDA100RT, SDB200RT), performance benchmarks, forward curves, token pricing datasets, and carbon footprint models, delivered via a self-serve web portal, REST API, and distribution through Bloomberg Terminal, Refinitiv, and Kaiko.

Is Silicon Data a public or private company?

Silicon Data is a private company. It is classified as venture growth investor backed and is currently operating.

When was Silicon Data founded?

Silicon Data was founded in 2024. It employs 11 to 50 people.

Where is Silicon Data based?

Silicon Data is headquartered in New York, United States, in the North America region.

How does Silicon Data make money?

Four revenue lines are on record. Silicon Data Portal Subscription is the primary driver. The others are tiered Data Subscriptions (SiliconNavigator™), index Licensing & Data Distribution and raw Data Delivery (GPU Pricing - Raw Data).

Who are Silicon Data's main competitors?

Direct peers on record are Compute Exchange, Kaiko, Cast.ai, CloudZero and MLPerf. Broad incumbents are CoreWeave, Crusoe Energy, Lambda Labs and Together AI. Fluidstack is listed as an emerging player.

Does Silicon Data have an API?

Yes. Silicon Data offers a REST API that enables developers to programmatically access GPU market intelligence data, including standardized indices (e.g., SDH100RT, SDB200RT), spot and reserved pricing, forward curves, regional pricing breakdowns, and the full historical archive. The API supports both portal/dashboard access and direct programmatic integration, with licensing options for internal use or product creation. Available as part of the Silicon Data Portal Access package ($499/month base) with full historical archive access. Developer documentation is at docs.silicondata.com/introduction.

Unlock the full company data

50 free credits on sign-up, no credit card required.

Contact sales
Live signals
CoinMarketCapWall Street puts a price on AI compute as miners pivot: Guest Post by Cryptopolitan_NewsCME Group will list AI compute futures on October 5, 2026, tracking Silicon Data's Nvidia index. Bitcoin miners have signed over $70 billion in AI compute contracts, and compute futures could reach $10 trillion in notional volume by 2030.Crypto BriefingSilicon Data CEO Carmen Li wants to turn GPU prices into a futures marketSilicon Data CEO Carmen Li is building pricing indices and a trading platform to hedge volatile AI compute costs, with CME Group planning cash-settled GPU futures and options. The company raised $35.2 million in funding, and CME targets a fall 2026 launch pending regulatory approval.HedgefundalphaSilicon Data Appoints Syntax Data As Independent Index Calculation Agent, Advancing Institutional Infrastructure For Compute Risk ManagementSyntax Data was appointed index calculation agent for Silicon Data's benchmark index family covering GPU and AI infrastructure markets. The appointment brings independent, third-party calculation to benchmarks that serve as reference rates for a new asset class.BenzingaDoes Michael Burry Have It Wrong? Nvidia Chips Retain Value, Data Shows - NVIDIA (NASDAQ:NVDA)Six-year-old Nvidia A100 chips retain nearly $5,000 residual value, complicating Michael Burry's warning that Big Tech overstays depreciation. Silicon Data estimates $4,956, with CoreWeave extending A100 contracts through 2029. Burry argues chips can rent and still depreciate fast.FinancialContent Business PageSilicon Data Appoints Syntax Data as Independent Index Calculation Agent, Advancing Institutional Infrastructure for Compute Risk ManagementSyntax Data was appointed index calculation agent for Silicon Data's benchmark index family covering GPU and AI infrastructure markets. The arrangement provides independent, third-party daily calculations under documented methodologies and audit oversight. Both firms expect the benchmark family to broaden as new hardware generations and contract tenors become established.MorningstarSilicon Data Appoints Syntax Data as Independent Index Calculation Agent, Advancing Institutional Infrastructure for Compute Risk ManagementSyntax Data was appointed as independent index calculation agent for Silicon Data's benchmark index family covering GPU and AI infrastructure markets. The arrangement provides daily calculations under documented methodologies and institutional governance, separating the calculation agent from the index sponsor. Both firms expect the benchmark family to broaden as new hardware generations and contract tenors become established.Business Wire BlogSilicon Data Appoints Syntax Data as Independent Index Calculation Agent, Advancing Institutional Infrastructure for Compute Risk ManagementSilicon Data appointed Syntax Data as its independent index calculation agent for GPU and AI infrastructure benchmarks. Syntax will perform daily calculations under documented methodologies and audit processes. The arrangement supports expansion of compute-linked products, including listed derivatives and structured solutions.TechStartupsAI Token Prices Collapse Nearly 60%, Raising Questions About the $600 Billion AI BoomSilicon Data's LLM token expenditure index fell to $0.9665 per million tokens on August 31, down nearly 60% from its May peak. The drop reflects competition and model routing to cheaper alternatives, not reduced demand. Q3 earnings will test whether consumption can outpace falling prices.PanewslabWhen computing power itself becomes a trading asset - from computing power futures to computing power dollarsCME Group and Silicon Data announced plans to launch computing power futures contracts, with two contracts on October 5 tracking H100 and B200 chip rental indices. The contracts, if successfully launched, will be listed on NYMEX.EIN PresswireSilicon Data Launches Monthly GPU Residual Value Benchmarks for AI Infrastructure FinanceSilicon Data launched monthly GPU residual value benchmarks for H100, A100, and B200, using a discounted cash flow model based on projected rental income. The estimates aim to provide fair-value guidance for asset-backed financing and fleet planning, complementing existing pricing indices.