← All companies & organizations

Chips / Company profile

Rebellions

What Rebellions does

Rebellions is a semiconductor and systems company focused on AI inference infrastructure—purpose-built accelerators and an accompanying software stack aimed at reducing the power and operational cost of running frontier models in data centers and enterprise environments. The company’s product portfolio spans inference NPUs (e.g., ATOM™ and the chiplet-based REBEL generation culminating in Rebel100™/Rebel-Quad), and rack-scale infrastructure systems (e.g., RebelServer™, RebelRack™, and RebelPOD™) that integrate compute, networking, management, and a software/serving layer designed for production deployment.

Rebellions’ technical differentiation is its end-to-end inference orientation: it pairs accelerator hardware optimized for inference with a compiler/runtime/software stack and Kubernetes/OpenShift-oriented deployment integrations. On the hardware side, Rebellions describes ATOM™ as a Samsung 5nm inference NPU with an architecture centered on neural engines and substantial on-chip SRAM, while its next-generation Rebel100™ is described as a peta-scale, UCIe-Advanced chiplet SoC approach designed to run mixed precision (FP16/FP8) efficiently with rack-scale performance claims. On the software side, Rebellions describes an SDK that includes a compiler toolchain and runtime, along with profiling/optimization capabilities, and an inference serving orientation that is explicitly tied to frameworks such as vLLM and Kubernetes/OpenShift compatibility.

Commercially, Rebellions positions its systems as deployable “AI infrastructure” rather than standalone chips, and it has been expanding partnerships across data center ecosystems and telecom/infrastructure operators. Recent activity includes moving further “up the stack” via acquisitions (e.g., SqueezeBits for inference optimization and serving), launching and scaling rack-scale product offerings, and announcing collaborations with large ecosystem players (e.g., Arm and SK Telecom) targeting sovereign AI / telecom-specific data center deployments.

News

Company record

Jul 24, 2026 · Official · RebellionsRebellions Successfully Runs SK Telecom’s Sovereign LLM ‘A.X K1’ on RebelServer, Delivering Both Efficiency and Real-World Viability

Rebellions announced a demonstration/press release claiming successful execution of SK Telecom’s A.X K1 on its RebelServer infrastructure and described an agent-service demo based on that setup.

↗
Jun 30, 2026 · Official · RebellionsRebellions Acquires AI Inference Optimization Leader SqueezeBits to Deliver Full-Stack AI Infrastructure

Rebellions announced the acquisition of SqueezeBits to bring inference optimization/serving capabilities in-house and evolve toward an integrated AI infrastructure platform spanning NPU hardware, software optimization, and inference serving.

↗
Jun 17, 2026 · Official · RebellionsRebellions and Giga Computing Sign MOU to Develop Next-Generation AI Server and Rack-Scale Solutions

Rebellions announced an MOU with Giga Computing to jointly develop next-generation AI server and rack-scale solutions based on Rebellions’ RebelCard™, aiming at sovereign AI and scalable infrastructure deployments.

↗
Apr 10, 2026 · Official · RebellionsRebellions Collaborates with SK Telecom and Arm Targeting Sovereign AI and Telecom Infrastructure

Rebellions announced a collaboration with SK Telecom and Arm, including integration efforts combining an Arm AGI CPU server concept with Rebellions’ RebelCard™, and stated plans to co-develop a software stack for sovereign AI workloads validated in SK Telecom’s data center environment.

↗
Mar 30, 2026 · Official · RebellionsRebellions Closes $400 Million Pre-IPO and Launches RebelRack™ and RebelPOD™ to Accelerate Global Expansion

Rebellions announced the $400M pre-IPO round and described the availability of RebelRack™ and RebelPOD™ as turnkey, vertically integrated rack-scale inference infrastructure offerings.

↗
Dec 11, 2025 · Official · RebellionsRebellions and Red Hat Introduce Red Hat OpenShift AI Powered by Rebellions NPUs to Fuel Choice and Flexibility in Enterprise AI

Rebellions and Red Hat announced OpenShift AI powered by Rebellions NPUs, positioning the solution as an enterprise-ready AI platform with native support for Rebellions software on Red Hat OpenShift AI.

↗
Nov 10, 2025 · Official · RebellionsRebellions Scales Global Growth with Silicon Valley backed Series C

Rebellions announced completion/expansion of its Series C with new participation from Kindred Ventures and Top Tier Capital Partners, describing U.S. market expansion momentum.

↗
Sep 30, 2025 · Official · RebellionsRebellions Raises $250 Million to Advance the Next Generation AI Infrastructure, Backed by Arm and Samsung

Rebellions announced its $250M Series C at a $1.4B valuation, naming Arm and Samsung-related strategic support and describing scaling of the company’s chiplet-based roadmap.

↗
Source map · 2 recurring channels · 54 references
RebellionsOfficial Site · monitor↗MTS Frontier / Control RoomCurated Feed · monitor↗discuss.rebellions.aiOfficial Reference · reference↗jp.rebellions.aiOfficial Reference · reference↗kr.rebellions.aiOfficial Reference · reference↗Official company sourceOfficial Reference · reference↗sec.govRegulatory · reference↗app.dealroom.coReporting · reference↗appext.hks.harvard.eduReporting · reference↗arxiv.orgReporting · reference↗axios.comReporting · reference↗bouncewatch.comReporting · reference↗cbinsights.comReporting · reference↗coasean.comReporting · reference↗congress.govReporting · reference↗crunchbase.comReporting · reference↗d18rn0p25nwr6d.cloudfront.netReporting · reference↗dealroom.coReporting · reference↗en.sedaily.comReporting · reference↗en.wikipedia.orgReporting · reference↗en.wowtale.netReporting · reference↗founderlodge.comReporting · reference↗gaebler.comReporting · reference↗goodrebellion.comReporting · reference↗govinfo.govReporting · reference↗inferencechips.comReporting · reference↗investor.xponential.comReporting · reference↗investors.predictive-oncology.comReporting · reference↗jonpeddie.comReporting · reference↗kisacoresearch.comReporting · reference↗linkedin.comReporting · reference↗market.vlsi.krReporting · reference↗marketscreener.comReporting · reference↗morenocapital.coReporting · reference↗neuronfeed.comReporting · reference↗polyn.aiReporting · reference↗reddit.comReporting · reference↗reportify-1252068037.cos.ap-beijing.myqcloud.comReporting · reference↗s201.q4cdn.comReporting · reference↗s204.q4cdn.comReporting · reference↗s205.q4cdn.comReporting · reference↗seedtable.comReporting · reference↗siliconvalleyinvestclub.comReporting · reference↗sipearl.comReporting · reference↗startup.genisisiq.comReporting · reference↗stocklight.comReporting · reference↗techcrunch.comReporting · reference↗techradar.comReporting · reference↗thenextweb.comReporting · reference↗theofficialboard.comReporting · reference↗theorg.comReporting · reference↗tomshardware.comReporting · reference↗vcletter.co.krReporting · reference↗webrazzi.comReporting · reference↗woodenscale.aiReporting · reference↗wsj.comReporting · reference↗

Still resolving: Newsroom · Official X

Funding

Latest disclosed valuation$2.3B
Tracked capital$1.1B6 sourced rounds

What it builds

ATOM™

ATOM™ is Rebellions’ inference-first AI accelerator described as manufactured on Samsung’s 5nm process and engineered for AI inference tasks, with performance and on-chip SRAM specifications stated in Rebellions’ developer materials.

source ↗
ATOM™-Max

ATOM™-Max is listed by Rebellions in its SDK support/compatibility information as a supported ATOM variant for inference workloads.

source ↗
Rebel100™

Rebel100™ is Rebellions’ rack-scale, peta-scale inference SoC based on a UCIe-Advanced chiplet architecture, described as executing mixed precision (FP8 and FP16) and using a predictive DMA coupled with an on-chip mesh to support high-throughput inference.

source ↗
RebelCard™

RebelCard™ is a module-type accelerator card described by Rebellions as integrating multiple NPU chiplets with HBM3E to deliver GPU-class performance with higher power efficiency for large-scale AI data centers.

source ↗
RebelServer™

RebelServer™ is Rebellions’ server platform for AI inference deployment; the company describes a rack/system family that integrates RebelServer, networking, and the Rebellions software stack into turnkey solutions.

source ↗
RebelRack™

RebelRack™ is Rebellions’ turnkey AI inference rack described as combining RebelServer, high-speed networking, and the Rebellions software stack into a validated rack-scale solution aimed at faster production readiness.

source ↗
RebelPOD™

RebelPOD™ is Rebellions’ pre-validated AI factory infrastructure built from integrated RebelRack units, described as delivering fixed configurations (e.g., 32/64/128/256 nodes) intended to simplify large-scale inference deployment.

source ↗
Rebellions Inference Platform

Rebellions Inference Platform is the company’s inference software/platform layer that Rebellions describes as purpose-built for inference hardware from a single accelerator card to rack-scale infrastructure, with stated support for Kubernetes and Red Hat OpenShift integration.

source ↗

Key metrics

Milestones & partnerships