← All companies

Chips / Company profile

FuriosaAI

What FuriosaAI does

FuriosaAI is a fabless semiconductor company building purpose-designed AI inference accelerators and the associated software toolchain to run frontier large language models (LLMs) and multimodal workloads in power- and cost-constrained enterprise data centers and cloud environments. The company’s flagship compute is RNGD (“Renegade”), a Tensor Contraction Processor (TCP)-based accelerator positioned for high-throughput inference with an emphasis on energy efficiency, air-cooled deployment targets, and integration into existing production inference stacks.

At the system level, Furiosa sells both accelerator hardware and turnkey/enterprise appliances, including the NXT RNGD Server, which is described as a “3kW inference appliance” designed to fit air-cooled racks and enable large-scale inference without requiring data center retrofits typical of higher-power GPU deployments. Furiosa also promotes a structured evaluation and deployment pathway (“Furiosa Access”) to help customers and partners integrate RNGD into their environments. On the software side, Furiosa’s developer experience is centered on its compiler/runtime/LLM stack distributed through Furiosa Docs and SDK releases, intended to map high-level model code onto its hardware and support production deployment patterns.

Strategically, Furiosa appears to be moving from hardware sampling toward wider commercialization and ecosystemization. In 2025 and 2026, Furiosa publicly described RNGD as entering mass production/shipping and announced multiple ecosystem and distribution partnerships, including integration via Microsoft’s Azure Marketplace, cloud-service enablement via Samsung SDS’s NPU-as-a-Service launch (for the Korean market), and a partnership with Broadcom to develop a next-generation inference platform built on Furiosa’s TCP approach and Broadcom’s XPU-related networking/fabric ecosystem.

Overall, Furiosa’s differentiation is the combination of (1) TCP/Tensor-contraction-first hardware design, (2) a compiler/runtime aimed at efficient deployment for real inference serving patterns (batching, streaming, multi-user concurrency), and (3) packaging into enterprise-ready appliances and cloud service offers rather than only selling raw accelerators.

News

Company record

Jul 20, 2026 · Official · FuriosaAIFuriosaAI and Samsung SDS Launch Korea’s First Domestic NPUaaS to Expand Enterprise AI Access

Samsung SDS launched an NPU-as-a-Service offering in Korea powered by Furiosa’s RNGD accelerator on Samsung Cloud Platform, integrated with cloud virtualization and deployable in configurable card counts for inference workloads.

May 27, 2026 · Official · FuriosaAIFuriosaAI partners with Broadcom to build next-generation inference platform for the Agentic Era

Furiosa announced a strategic partnership with Broadcom to develop a next-generation inference platform that evolves Furiosa’s TCP architecture and ties it to Broadcom’s XPU technology and platform for large-scale agentic AI serving.

May 13, 2026 · Official · FuriosaAIExperience RENEGADE Summit 2026

Furiosa’s inaugural RENEGADE Summit recap highlights RNGD as its agentic-era inference engine and features a keynote by CEO June Paik.

Apr 02, 2026 · Official · FuriosaAIRENEGADE 2026 Summit: Key announcements and highlights from our global partner ecosystem

Furiosa recapped partner announcements around RNGD-powered products and services, including cloud and appliance-related collaborations.

Jan 29, 2026 · Official · FuriosaAI NewsroomFuriosaAI starts shipping (referenced in newsroom and covered by external press)

Furiosa’s newsroom links to a shipping announcement covered by external media; the primary shipping details in this profile are sourced from the mass-production post dated 2026-01-27.

Jan 27, 2026 · Official · FuriosaAIRNGD enters mass production (4,000 High-Performance AI Accelerators Shipped by TSMC)

Furiosa stated it is officially shipping RNGD in volume and described delivery of an initial batch of 4,000 units from partners including TSMC and ASUS, with RNGD available as a PCIe card and as a turnkey server.

Jul 30, 2025 · Official · FuriosaAIAnnouncing FuriosaAI’s $125M Series C bridge funding to scale sustainable AI compute

Furiosa closed a $125M Series C bridge round, citing RNGD demand from enterprise customers and a roadmap acceleration for sustainable AI compute.

Apr 16, 2025 · Official · FuriosaAIFuriosa to bring RNGD to Microsoft’s Azure Marketplace

Furiosa announced RNGD availability on Microsoft’s Azure Marketplace, describing streamlined deployment for enterprise customers and plans for inference APIs based on precompiled Llama models.

Show 1 earlier updates
Source map · 2 recurring channels · 18 references

Still resolving: Newsroom · Official X

Funding

Latest disclosed valuationNot disclosed
Tracked capital$800.2B7 sourced rounds
DateRoundRaisedValuationLead / investorsEvidence
May 28, 2026Equity investment (National Growth Fund)$800B
Not attributed
+1 investor
May 31, 2021Series B (amount not disclosed in accessible source)
Apr 01, 2017Seed
Not attributed
+1 investor
UndatedSeed / early investment (year and amount not fully disclosed in accessible sources)

What it builds

RNGD (“Renegade”)

RNGD is FuriosaAI’s flagship TCP-based inference accelerator targeted for high-performance LLM and multimodal inference, with an emphasis on power efficiency and air-cooled data-center operation. The product page positions RNGD as delivering high performance while maintaining a low power profile and supports PCIe-based deployment.

source ↗
NXT RNGD Server

NXT RNGD Server is FuriosaAI’s enterprise-ready, multi-accelerator inference appliance intended for on-premises and cloud deployments. Furiosa describes it as an air-cooled, turnkey server designed for high-throughput inference for advanced LLM and agentic AI workloads.

source ↗
Furiosa SDK / Furiosa software stack (developer tooling)

Furiosa markets a software stack for compiling, optimizing, and deploying LLM inference and agentic workloads on Furiosa hardware, with documentation hosted at Furiosa Developer Center and SDK releases referenced across Furiosa news and product materials.

source ↗

Key metrics

Milestones & partnerships

Jul 20, 2026
Samsung SDS NPUaaS launch in Korea powered by RNGD

Samsung SDS launched a domestically offered NPU-as-a-Service built on RNGD on Samsung Cloud Platform.

Samsung SDS
May 27, 2026
Broadcom partnership for next-generation inference platform

Furiosa announced a partnership with Broadcom to evolve TCP into a scale-up inference platform for agentic-era serving.

Broadcom
Apr 02, 2026
Partner ecosystem announcements at RENEGADE 2026

Furiosa summarized partner ecosystem announcements for RNGD-powered services and infrastructure offerings.

Jan 27, 2026
RNGD mass production/shipping in volume

Furiosa stated it is shipping RNGD in volume and described initial unit deliveries from TSMC and ASUS, with availability as both PCIe cards and turnkey servers.

TSMC · ASUS
Jul 30, 2025
Series C bridge funding to scale sustainable AI compute

Furiosa announced a $125M Series C bridge round to support RNGD demand and roadmap execution.

Korea Development Bank · Industrial Bank of Korea · Keistone Partners · PI Partners · Kakao Investment
Apr 16, 2025
RNGD integration on Microsoft Azure Marketplace

Furiosa announced the launch of RNGD as a product offering on Azure Marketplace and described deployment and scaling via Azure.

Microsoft
Dec 19, 2024
RNGD performance and SDK roadmap updates for Llama 3.1

Furiosa reported throughput and single-user performance figures for Llama 3.1 workloads on RNGD and discussed planned SDK v2024.3.0 capabilities.

Apr 04, 2017
Company establishment date (Korea registry listing)

A Korean company information listing shows FuriosaAI’s establishment date as 2017-04-04.