Close Menu
TechTost
  • AI
  • Apps
  • Crypto
  • Fintech
  • Hardware
  • Media & Entertainment
  • Security
  • Startups
  • Transportation
  • Venture
  • Recommended Essentials
What's Hot

Cognition’s Scott Wu says AI coding agents shouldn’t replace humans

Slate Auto will announce pricing and take pre-orders for its EV on June 24

In just 3 weeks, StrictlyVC is coming to Los Angeles

Facebook X (Twitter) Instagram
  • About Us
  • Contact Us
  • Privacy Policy
  • Terms and Conditions
  • Disclaimer
Facebook X (Twitter) Instagram
TechTost
Subscribe Now
  • AI

    This chip startup just raised $135 million on a bet that AI’s biggest bottleneck isn’t computation — it’s memory

    29 May 2026

    Glean’s top line tops $300M as AI budget cut becomes its main selling point

    29 May 2026

    How long is Anthropic’s lease with SpaceX? Opinions vary.

    28 May 2026

    Why Google’s AI Can’t Type Google (or Anything)

    28 May 2026

    ElevenLabs’ new music generation model can switch genres mid-track

    27 May 2026
  • Apps

    A sneak peek at the new Siri app reveals Apple’s plans to tackle ChatGPT and more

    29 May 2026

    Bluesky embraces long-form content to tackle X articles

    29 May 2026

    Sesame, the AI ​​chat startup from the founders of Oculus, is launching its iOS app

    28 May 2026

    Airbnb-backed WeRoad raises $58 million to bring its group travel platform to the US

    28 May 2026

    Spotify now lets you “clip” moments from your favorite podcast

    27 May 2026
  • Crypto

    Startup Battlefield 200 applications close today

    27 May 2026

    5 days left: Save up to $410 on Disrupt 2026 passes

    25 May 2026

    As crypto cools, a16z crypto raises $2.2 billion in capital

    6 May 2026

    Coinbase to lay off 14% of staff as part of broader restructuring

    5 May 2026

    British cryptographer Adam Back denies NYT report that he is Bitcoin creator Satoshi Nakamoto

    9 April 2026
  • Fintech

    Last 24 hours to save up to $410 on your Disrupt 2026 ticket

    29 May 2026

    2 days left: Lock in up to $410 in ticket savings for Disrupt 2026

    28 May 2026

    Robinhood now allows your AI agents to trade stocks

    28 May 2026

    Disrupt 2026 Early Bird ticket savings expire in 3 days

    27 May 2026

    Disrupt 2026 Early Bird ticket prices end May 29

    26 May 2026
  • Hardware

    Kiwibit’s artificial intelligence bird feeder is my new backyard friend

    29 May 2026

    Vertu wants CEOs to run companies from a foldable AI starting at $6,880

    29 May 2026

    Oura unveils its Ring 5 with a thinner, lighter design starting at $399

    28 May 2026

    The Dreamie alarm clock made me stop using my phone in bed

    26 May 2026

    6 kitchen gadgets that make adult life easier

    25 May 2026
  • Media & Entertainment

    YouTube will automatically flag videos with artificial intelligence

    28 May 2026

    Meta launches Instagram, Facebook and WhatsApp subscriptions, with more to follow, including AI plans

    27 May 2026

    Spotify now lets you view narrated magazine articles as well

    26 May 2026

    Spotify launches an audiobook creation tool powered by ElevenLabs

    22 May 2026

    New York City Mayor Zohran Mamdani Takes To Twitch To Chat With New Yorkers

    21 May 2026
  • Security

    A security flaw in prison payphone service Pay Tel exposed publicly the driver’s licenses of more than 300,000 callers

    29 May 2026

    Hackers are trying to steal Signal users’ backups in new wave of phishing attacks

    28 May 2026

    CrowdStrike and Google take down botnet used by hackers to target open source software developers

    28 May 2026

    UK Visa Portal Revealed Thousands of Applicants’ Passports and Selfies — Then Invited Lawyers to Ask Us

    27 May 2026

    UK Visa portal leaked thousands of applicant passports and selfies online – and hasn’t fixed the leak

    27 May 2026
  • Startups

    Cognition’s Scott Wu says AI coding agents shouldn’t replace humans

    29 May 2026

    How to apply to Startup Battlefield 2026, what you need before the June 8 deadline

    29 May 2026

    At Disrupt 2026: Databricks co-founder on what’s killing AI business deals

    28 May 2026

    Tech CEOs apparently suffer from AI psychosis

    28 May 2026

    SOND, a sleep tech startup from former Bose sleep chief, exits stealth with $7 million

    27 May 2026
  • Transportation

    Slate Auto will announce pricing and take pre-orders for its EV on June 24

    29 May 2026

    Waymo dominates autonomous vehicle registrations as Tesla follows

    29 May 2026

    Slate Auto will begin taking orders for its affordable EV on June 24

    28 May 2026

    FAA orders SpaceX to investigate Starship V3 booster failure

    27 May 2026

    The Trump administration is allowing Volvo to continue selling connected cars in the US

    27 May 2026
  • Venture

    In just 3 weeks, StrictlyVC is coming to Los Angeles

    29 May 2026

    Why Paris might be the most important AI city outside of Silicon Valley

    29 May 2026

    ClickHouse triples annual revenue to $250 million, charting a path to an IPO

    28 May 2026

    Triomics raises $22 million to bring oncology AI to cancer centers

    28 May 2026

    ClickHouse triples annual revenue to $250 million, charting a path to an IPO

    27 May 2026
  • Recommended Essentials
TechTost
You are at:Home»AI»This chip startup just raised $135 million on a bet that AI’s biggest bottleneck isn’t computation — it’s memory
AI

This chip startup just raised $135 million on a bet that AI’s biggest bottleneck isn’t computation — it’s memory

techtost.comBy techtost.com29 May 202604 Mins Read
Share Facebook Twitter Pinterest LinkedIn Tumblr Email
This Chip Startup Just Raised $135 Million On A Bet
Share
Facebook Twitter LinkedIn Pinterest Email

Every time you ask a question in ChatGPT, your request triggers a data relay race. Information leaves memory, passes through a CPU for preprocessing, travels to a GPU for heavy calculations, and then returns — and the entire journey is repeated for every word the AI ​​generates.

The bottleneck is structural — it means routing some of the most expensive and power-hungry chips in the industry for each request. This inefficiency is just that XCENAa startup with offices in South Korea and the US, is trying to solve. The four-year-old startup has designed a chip that puts computing capabilities much closer to DRAM — the fast, short-term memory chips that store data that the processor is actively using — allowing common data operations to be handled close to memory, without the costly round-trips between CPU, GPU and memory.

If it works at scale, the implications for AI infrastructure costs could be significant, which largely explains the investor excitement around the company. Indeed, XCENA just raised $135 million in a Series B at a $570 million valuation, bringing its total to $185 million.

XCENA CEO Jin Kim co-founded the startup in 2022 with CTO Dohun Kim and CPO Harry Juhyun Kim, veterans of Samsung and SK Hynix, the memory giants that supply chips that power Nvidia’s GPUs. “CPUs and GPUs have gotten smarter over the decades. Memory never did. XCENA wants to change that,” Jin Kim told TechCrunch. “The recent rise in memory prices and related stocks indicates a broader shift in AI infrastructure toward memory-centric architectures,” he added. (This month, the three companies that dominate the global memory chip market — Samsung, SK Hynix and Micron — each crossed a trillion-dollar valuation for the first time.)

XCENA is betting on the premise that “inference is not just a computational problem; it’s increasingly a memory-scale problem,” Kim said.

XCENA’s chip, the MX1, connects to the CPU via CXL (Compute Express Link) – essentially a dedicated express lane between the processor and memory – processing data before it has to leave the memory module. It brings computation to data, not the other way around. The company claims that what used to require 10 servers could potentially be run on just one.

“While GPUs excel at matrix multiplication—the heavy math behind AI model training—much of the data orchestration that surrounds it, including preprocessing, KV cache management (the system that stores the previous conversation frame so a model doesn’t have to process it again), and data caching, is still performed on a CPU. Our chip handles these tasks directly within the memory module itself.

Demand for memory solutions has picked up since the second half of last year, and the company believes the timing is working in its favor.

Talks with several global memory suppliers are in the early stages, although Kim declined to name them. The company’s ideal customers are hyperscalers who spend tens of billions a year on AI infrastructure, where even a small gain in memory performance can mean hundreds of millions in savings.

The MX1 is still a prototype. Mass-produced chips are scheduled to roll off Samsung’s foundry lines by the end of 2026, with the company expecting to generate revenue from 2027.

While neural processing unit (NPU) manufacturers compete to challenge Nvidia for training workloads, XCENA targets the memory-intensive layer that lies beneath it all.

XCENA’s closest rivals include Astera Labs and Marvell, Nasdaq-listed companies working on next-generation memory connectivity. Marvell is a large, established player already working in the same space, Kim said, adding that the difference is intellectual property. “We have thousands of cores,” Kim said. Based on the public specification, Marvell’s approach relies on a handful of general-purpose cores by comparison.

These cores are built on RISC-V – an open source chip design scheme – and optimized specifically for data processing, with each core intentionally kept small and efficient. Beyond the cores themselves, XCENA designs its own internal memory, bus interface and DRAM controller hierarchy — a level of vertical integration that most chip companies, including larger rivals, typically outsource.

Seoul-based VC firms Atinum and IMM Investment co-led the Series B round, along with Corstone Asia and existing investors SBI Investment and Mirae Asset Capital. The company, which has more than 90 employees in offices in Pangyo, a tech hub outside Seoul, and in Sunnyvale, is also in talks with international investors about additional funding.

When you purchase through links in our articles, we may earn a small commission. This does not affect our editorial independence.

AIs bet biggest bottleneck chip computation dram isnt lack of memory chip memory memory chip million raised Samsung Electronics SK Hynix South Korea startup XCENA
Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
Previous ArticleA sneak peek at the new Siri app reveals Apple’s plans to tackle ChatGPT and more
Next Article In just 3 weeks, StrictlyVC is coming to Los Angeles
bhanuprakash.cg
techtost.com
  • Website

Related Posts

How to apply to Startup Battlefield 2026, what you need before the June 8 deadline

29 May 2026

Glean’s top line tops $300M as AI budget cut becomes its main selling point

29 May 2026

ClickHouse triples annual revenue to $250 million, charting a path to an IPO

28 May 2026
Add A Comment

Leave A Reply Cancel Reply

Don't Miss

Cognition’s Scott Wu says AI coding agents shouldn’t replace humans

29 May 2026

Slate Auto will announce pricing and take pre-orders for its EV on June 24

29 May 2026

In just 3 weeks, StrictlyVC is coming to Los Angeles

29 May 2026
Stay In Touch
  • Facebook
  • YouTube
  • TikTok
  • WhatsApp
  • Twitter
  • Instagram
Fintech

Last 24 hours to save up to $410 on your Disrupt 2026 ticket

29 May 2026

2 days left: Lock in up to $410 in ticket savings for Disrupt 2026

28 May 2026

Robinhood now allows your AI agents to trade stocks

28 May 2026
Startups

Cognition’s Scott Wu says AI coding agents shouldn’t replace humans

How to apply to Startup Battlefield 2026, what you need before the June 8 deadline

At Disrupt 2026: Databricks co-founder on what’s killing AI business deals

© 2026 TechTost. All Rights Reserved
  • About Us
  • Contact Us
  • Privacy Policy
  • Terms and Conditions
  • Disclaimer

Type above and press Enter to search. Press Esc to cancel.