GigaIO
GigaIO builds Gryf, a portable carry-on-sized AI supercomputer co-designed with SourceCode, bringing datacenter-class inference to edge environments for defense, intelligence, energy, media, and sports analytics customers, sold via direct sales and channel partners.
- Company typePrivate
- Founded2019
- HeadquartersCarlsbad, United States
- Headcount11–50
- GTM typeB2B
- OfferingHardware or Manufacturing
What GigaIO does
GigaIO (legal entity: GigaIO Networks, Inc.) is a venture-backed private company headquartered in Carlsbad, California, that develops portable edge AI infrastructure. The firm's flagship product, Gryf, is described as the world's first carry-on suitcase-sized AI supercomputer co-designed with manufacturing partner SourceCode. Gryf is built on a modular sled architecture offering configurable GPU, compute, storage, and networking options with up to 2,500W power capacity and support for up to 64 datacenter-class GPUs in a single system. The platform enables inference workloads including agentic AI, generative AI, machine learning, deep learning, and ISR analytics at the edge without cloud dependency, and up to five units can be stacked in an all-to-all topology for petabyte-scale processing.
Underlying technology centers on composable infrastructure that disaggregates and re-aggregates compute, memory, storage, and networking resources at the rack or cluster level using PCIe Gen 5. Historically the company also developed the FabreX AI fabric and SuperNODE rack-scale platform, both of which were sold to d-Matrix in April 2026. Following this divestiture, GigaIO has refocused exclusively on the portable edge Gryf business, transferring the data-center engineering team with those assets to d-Matrix.
GigaIO monetizes primarily through quote-based hardware sales of Gryf systems sold via direct enterprise field sales augmented by a channel partner/reseller network (Dell, Supermicro, WWT). Pricing is bundled across multi-year contracts with custom configurations for defense, intelligence, energy, media/broadcasting, and sports analytics customers. Logo customer concentration skews heavily toward U.S. federal research and defense (DoD, Argonne, Lawrence Livermore, Oak Ridge, MITRE) with select enterprise and channel partners (Morgan Stanley, NVIDIA, Ericsson, IBM, VISA, NTT, Fujitsu, Oxford Nanopore). Total disclosed equity raised across 2018, 2021, and 2025 rounds exceeds $40M, with the most recent $21M Series B first tranche closing in July 2025 to expand Gryf production and accelerate AI inferencing innovation.
GigaIO firmographics
Firmographics- Name
- GigaIO
- Legal name
- GigaIO Networks, Inc.
- Website
- https://gigaio.com
- Company type
- Private
- Founded year
- 2019
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Short description
- GigaIO builds Gryf, a portable carry-on-sized AI supercomputer co-designed with SourceCode, bringing datacenter-class inference to edge environments for defense, intelligence, energy, media, and sports analytics customers, sold via direct sales and channel partners.
- Ownership category
- akta.pro rank
GigaIO industry classification
Industry- Product category
- Edge AI Infrastructure Hardware
- NAICS
- Computer and Peripheral Equipment Manufacturing (33411), Computer Systems Design and Related Services (54151), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182)
- SIC
- Electronic Computers (3571), Services-Computer Integrated Systems Design (7373), Computer Communications Equipment (3576)
- akta.pro primary industry
- Edge AI Hardware & NPUs (SoCs, modules, accelerators) (HDAAAJAC)
- akta.pro secondary industries
- Model Serving, Inference & Deployment Platforms (APIs, Edge/On-Prem) (HDAEANAC), Real-Time Vision at the Edge (on-device CV/video) (HDAAAJAE)
Keywords
Where GigaIO is headquartered
LocationHeadquarters
- HQ city
- Carlsbad
- HQ country
- United States
- HQ region
- North America
Offices1 record
Markets served
GigaIO business model
Business model- GTM type
- B2B
- Offering type
- Hardware or Manufacturing
- Cost components
- Technology or R&D, Personnel, Supply Chain, Marketing or Sales, Operations
Revenue model
- Gryf Hardware Sales: Sale of Gryf portable AI supercomputer systems and accessories. The product is sold as a complete hardware system with configurable sled options for GPU, compute, storage, and networking. Revenue recognized upon hardware delivery with potential professional services for deployment and integration.
- Enterprise Software and Support Services: Software licensing for FabreX technology (prior to sale to d-Matrix), software updates, technical support, and professional services for system deployment and optimization.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Unit Pricing | Multi-year contract | Gryf Base Configuration |
| Other | Multi-year contract | Enterprise/Federal Custom Solutions |
Go-to-market motion2 records
Distribution channels2 records
Marketing channels5 records
GigaIO product offering
Product offeringCore offering
GigaIO designs, manufactures, and sells portable edge AI infrastructure, with its flagship product Gryf being the world's first carry-on suitcase-sized AI supercomputer co-designed with SourceCode. Following the April 2026 sale of its data center business (SuperNODE and FabreX) to d-Matrix, GigaIO focuses exclusively on delivering datacenter-class AI inferencing in a ruggedized, field-deployable form factor for defense, intelligence, energy, media, and sports analytics customers. The product features modular sled architecture with configurable GPU, compute, storage, and networking options.
Product overview
GigaIO delivers datacenter-class computing at the edge through its flagship product Gryf, the world's first carry-on suitcase-sized AI supercomputer co-designed with SourceCode. Following the April 2026 sale of its data center business (SuperNODE and FabreX) to d-Matrix, GigaIO now focuses exclusively on portable edge AI infrastructure. Gryf provides modular, field-serviceable AI inferencing with up to 64 datacenter-class GPUs, supporting agentic AI, machine/deep learning, and ISR analytics across defense, energy, media, and sports sectors. The platform features patented composable infrastructure enabling dynamic reconfiguration of compute, accelerator, storage, and network sleds without hardware changes.
Differentiator
Problem solved
Functional benefit
Products and services
- Gryf Portable AI Supercomputer The world's first carry-on suitcase-sized AI supercomputer co-designed with SourceCode, bringing datacenter-class computing power directly to edge locations. Features modular sled architecture with configurable GPU, compute, storage, and networking options, supports up to 2,500W power capacity (up to 10x competitor capability), up to 64 datacenter-class GPUs in a single system, and up to 5 units can be stacked in an all-to-all topology for petabyte-scale processing. Designed for defense, intelligence, sports, media, and energy sectors requiring AI inferencing in environments with poor or no connectivity.
Quantifiable outcome
- Up to 10x compute capability compared to competitors offering only 250-1,500W power capacity
- +3 more outcomes
Companies that use GigaIO
Customer profileNamed customers11 records
Segments4 records
Ideal customer profiles2 records
GigaIO technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- No
- API docs
- API detail
Core technology
AI maturity
App detail
AI capability8 records
Feature4 records
GigaIO partnerships and signals
Strategic signalPartnerships
Ten partnerships are on record, tiered supporting and core.
- EdgeRunner AIsupportingDelivered agentic generative AI at the edge for enterprise, media, and defense applications. Co-design engineering effort combining Gryf hardware with EdgeRunner AI software for edge GenAI capabilities.
- SourceCodecoreCo-design partner for Gryf portable AI supercomputer. SourceCode provides ruggedized system design and co-design manufacturing capabilities. Partnership combines GigaIO's interconnect and composable infrastructure expertise with SourceCode's hardware engineering.
- HammerspacesupportingPartnership enables Hammerspace Data Orchestration platform to process, store, and orchestrate data in edge compute environments using Gryf. Combines data orchestration with portable GPU computing for distributed edge AI workloads.
- NVIDIAcorePremier member of NVIDIA Inception program. Technology ecosystem partner supporting NVIDIA GPUs in Gryf configurations. Open platform supports NVIDIA accelerators alongside other GPU vendors.
- AMDsupportingTechnology ecosystem partner. Gryf supports AMD GPUs as part of open accelerator strategy enabling customers to choose best hardware for their workloads.
- BroadcomsupportingTechnology ecosystem partner. Previously collaborated with d-Matrix on SquadRack reference architecture. Gryf platform supports Broadcom networking components.
- MicronsupportingMemory technology partner in GigaIO ecosystem. Supports high-performance memory configurations for AI workloads in Gryf and SuperNODE platforms.
- SupermicrosupportingListed partner offering server and storage solutions. Both companies serve HPC and AI infrastructure markets with complementary product lines.
- Dell TechnologiessupportingListed partner and customer. Both companies serve enterprise and government markets with HPC and AI infrastructure solutions.
- World Wide Technologies (WWT)supportingMajor technology solutions provider and channel partner. WWT provides advanced IT infrastructure solutions and serves government and enterprise customers.
Scale indicators3 records
Recent moves7 records
Expansion highlights4 records
GigaIO competitors and assessment
Company assessmentDirect peers
- Groq: Groq produces LPU-based AI inference appliances emphasizing low-latency, high-throughput inferencing — overlapping GigaIO's positioning for datacenter-class inference performance at the edge.
- SambaNova Systems: SambaNova builds AI inference accelerator systems targeting datacenter and edge deployments, competing head-on with GigaIO's composable inference architecture for high-performance enterprise AI workloads.
- Cerebras Systems: Cerebras builds wafer-scale AI compute systems for training and inference, targeting enterprise and government AI workloads in a comparable high-performance, specialized-hardware category.
Broad incumbents
- NVIDIA: NVIDIA's Jetson and IGX edge AI platforms compete in portable edge AI inference, and the company is both a GigaIO partner (Premier Inception member) and a potential competitor with vastly greater R&D resources.
- Supermicro: Supermicro offers a broad portfolio of GPU-dense servers and edge AI systems, competes in HPC and AI infrastructure, and is also a GigaIO channel partner.
- Dell Technologies: Dell sells ruggedized PowerEdge XR servers and HPC solutions targeted at defense, federal, and edge AI deployments; it is both a GigaIO customer/partner and a broad incumbent offering portable enterprise compute.
- Hewlett Packard Enterprise (HPE): HPE's Edgeline and Cray HPC systems target industrial edge and AI inferencing in remote and rugged environments, directly overlapping GigaIO's defense, energy, and remote operations use cases.
Emerging players
- Positron: Positron builds memory-optimized AI inference accelerators targeting datacenter and edge deployments, competing with GigaIO's composable inference architecture for low-latency, high-throughput inference workloads.
- d-Matrix: d-Matrix acquired GigaIO's SuperNODE and FabreX assets in April 2026 and now competes in datacenter-scale AI inference — adjacent to GigaIO's remaining edge AI focus.
- Lambda Labs: Lambda designs GPU-dense AI training and inference systems (cloud and on-prem), competing with GigaIO in the high-performance AI infrastructure category with a similar enterprise and AI-developer customer profile.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat4 records
Key risks6 records
Key highlights7 records
Customer concentration
GigaIO social profiles
Digital presenceGigaIO financial estimates
Financial estimateRevenue estimate
Valuation estimate
GigaIO leadership team
Management profileNumber of profiles
Profiles5 records
GigaIO funding detail
Funding detailFunding overview
Funding rounds4 records
Investors10 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
GigaIO M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about GigaIO
What does GigaIO do?
GigaIO designs, manufactures, and sells portable edge AI infrastructure, with its flagship product Gryf being the world's first carry-on suitcase-sized AI supercomputer co-designed with SourceCode. Following the April 2026 sale of its data center business (SuperNODE and FabreX) to d-Matrix, GigaIO focuses exclusively on delivering datacenter-class AI inferencing in a ruggedized, field-deployable form factor for defense, intelligence, energy, media, and sports analytics customers. The product features modular sled architecture with configurable GPU, compute, storage, and networking options.
Is GigaIO a public or private company?
GigaIO is a private company. It is classified as venture growth investor backed and is currently operating.
When was GigaIO founded?
GigaIO was founded in 2019. It employs 11 to 50 people.
Where is GigaIO based?
GigaIO is headquartered in Carlsbad, United States, in the North America region.
How does GigaIO make money?
Two revenue lines are on record. Gryf Hardware Sales are the primary driver. The others are enterprise Software and Support Services.
Who are GigaIO's main competitors?
Direct peers on record are Groq, SambaNova Systems and Cerebras Systems. Broad incumbents are NVIDIA, Supermicro, Dell Technologies and Hewlett Packard Enterprise (HPE). Emerging players are Positron, d-Matrix and Lambda Labs.
Does GigaIO have an API?
No public API is recorded for GigaIO.
What industry is GigaIO in?
GigaIO's product category is Edge AI Infrastructure Hardware. Its primary akta.pro industry code is HDAAAJAC, Edge AI Hardware & NPUs (SoCs, modules, accelerators), with a secondary code of HDAEANAC, Model Serving, Inference & Deployment Platforms (APIs, Edge/On-Prem). Its NAICS code is 33411 and its SIC code is 3571.