Work Here?
Overview
MatX builds high-throughput AI accelerators for large language models. Its main product is a dedicated chip where every transistor is optimized for running large neural networks, aiming to deliver strong performance for training and inference of LLMs. The company also develops the surrounding software stack, including compilers, to make the hardware easy to program for ML engineers (e.g., through JAX-based workflows). Unlike general-purpose processors, MatX focuses on specialization and full-stack optimization from silicon to software. Its goal is to enable organizations with substantial AI workloads to scale efficiently by providing powerful, purpose-built hardware and the compatible software tools they need to deploy and optimize large models.
About MatX
Industries
Hardware
AI & Machine Learning
Company Size
51-200
Company Stage
Series B
Total Funding
$604.9M
Headquarters
Mountain View, California
Founded
2022
Get referred to MatX
See people who can refer or advise you
Simplify's Take
What believers are saying
- MatX raised $500 million on February 24, 2026, from Jane Street and Situational Awareness.
- MatX says tapeout lands within a year, with shipping targeted for 2027.
- Greenhouse listed 30 open roles on July 30, 2026, signaling active scaling.
What critics are saying
- MatX ships in 2027, after Nvidia Rubin reaches partners in 2026.
- MatX sacrifices small-model performance and programming ease for LLM specialization.
- Any TSMC delay or failed tapeout pushes revenue beyond 2027, burning $500 million.
What makes MatX unique
- Founded by ex-Google TPU engineers Reiner Pope and Mike Gunter.
- MatX One targets LLM training and inference with split systolic arrays.
- MatX integrates silicon, compilers, kernels, and rack systems in-house.
Help us improve and share your feedback! Did you find this helpful?
Funding
Total Funding
$604.9M
Above
Industry Average
Funded Over
3 Rounds
Industry standards
Benefits
Company Equity
Growth & Insights and Company News
6 month growth
↓ -4%1 year growth
↓ -7%2 year growth
↑ 1%MatX, an LLM-focused chip startup, has raised $500 million in a Series B led by Jane Street and Situational Awareness LP, with participation from Spark Capital, the Collison brothers, Andrej Karpathy, Alchip and Marvell. The company is now valued at an undisclosed amount. The startup is developing the MatX One, a purpose-built AI inference chip targeting large language model workloads. The chip features a splittable systolic array architecture designed to maximise throughput and latency at scale, specifically for high-volume LLM inference. MatX explicitly sacrifices small-model performance and programming ease for specialised LLM performance. The company, now at 100 employees, expects tapeout within a year. The funding will accelerate development and manufacturing, with volume production targeted for the second half of 2027.
MatX raises $500 million to build AI training chips. The companu is building custom AI processors for training LLMs faster and more efficiently PUBLISHED: Mon, Mar 2, 2026, 11:14 AM UTC | UPDATED: Sat, Mar 7, 2026, 4:40 AM UTC 2 mins read Photo: Bloomberg * | MatX raised $500M in Series B funding to build custom AI processors optimized for large language model training and inference * | The company is designing specialized AI chips targeting roughly 10x better performance than current leading GPUs by optimizing hardware for matrix math and parallel AI workloads * | Founded by former Google TPU engineers, MatX plans to manufacture chips via TSMC with commercial shipping targeted for 2027 MatX has raised $500 million in Series B funding, led by Jane Street and Situational Awareness, a fund created by former Leopold Aschenbrenner. The startup is building custom AI processors designed specifically for training large language models faster and more efficiently than general-purpose GPUs. MatX's core ambition is aggressive: make its chips roughly 10x better at LLM training performance and inference efficiency compared to current leading hardware. The company is part of a broader shift in AI infrastructure, where specialized chips are becoming as important as models themselves. Instead of relying on general-purpose computing hardware, MatX is designing silicon optimized for matrix math, parallel AI workloads, and long-context model training. In practice, that means tailoring chip architecture, memory access patterns, and data movement efficiency specifically for AI workloads rather than traditional computing tasks. MatX was founded in 2023 by former Google TPU engineers. CEO Reiner Pope previously led AI software development for Google's Tensor Processing Units, while co-founder Mike Gunter helped design TPU hardware. Their experience shows in the company's approach: vertically integrating hardware and AI workload optimization rather than treating chips as generic computing components. The new capital will be used to manufacture chips through TSMC, with production and commercial shipping targeted for 2027. MatX is entering a highly competitive AI chip race against other startups and incumbents building alternatives to GPUs, as demand for AI training compute continues to outpace supply. More Topics:
The startup was founded by former Google TPU engineers in 2023.
Nvidia challenger AI chip startup MatX raised $500M. February 25, 2026 MatX, a chip startup founded by two former Google hardware engineers, has raised a $500 million Series B led by Jane Street and Situational Awareness, an investment fund formed by former OpenAI researcher Leopold Aschenbrenner. The company's goal is to make its processors 10 times better at training LLMs and delivering results than Nvidia's GPUs. Other investors in the round include Marvell Technology, NFDG, Spark Capital, and Stripe co-founders Patrick Collison and John Collison, the startup's founder and CEO Reiner Pope announced Tuesday in a post on LinkedIn. Although the company didn't release its latest valuation, Etched, MatX's closest competitor, raised a $500 million round at a $5 billion valuation, Bloomberg reported last month. Etched didn't immediately respond to a request for comment. MatX's latest round comes more than a year after its Series A of about $100 million, which was led by Spark Capital. TechCrunch earlier reported that the 2024 round valued the startup at more than $300 million. Before co-founding MatX in 2023, Pope led AI software development for Google's TPUs, the tech giant's proprietary AI chips. His co-founder, Mike Gunter, was a lead designer of the TPU hardware before leaving to launch the startup. The new funding will help MatX produce its chips with TSMC, with plans to start shipping them in 2027. Techcrunch event Boston, MA | June 9, 2026 #Nvidia #challenger #chip #startup #MatX #raised #500M February 25, 2026 Keep up to date with the most important news. Add a comment Add a comment Apple's new age verification instruments block underage app downloads the place required by legislation. Marking 4 years of struggle in ukraine with highly effective music and movies. February 25, 2026
US AI chip startup MatX raises $500m to compete with Nvidia. United States-based AI chip startup MatX has raised over US$500 million in a funding round led by Jane Street and Situational Awareness, a firm founded by ex-OpenAI researcher Leopold Aschenbrenner. The company, founded by former Google semiconductor engineers Reiner Pope and Mike Gunter, aims to develop hardware to compete with Nvidia in the AI chip market. MatX plans to finalize its chip design this year and begin shipping in 2027, working with Taiwan Semiconductor Manufacturing Co. (TSMC) to produce the hardware. The startup claims its chip can outperform Nvidia's upcoming Rubin Ultra based on performance per square millimeter, targeting AI training and inference workloads. Food for thought. Implications, context, and why it matters. ### MatX bets on a 'post-ChatGPT' design backed by an AGI-driven investment thesis * MatX started after massive transformer models (a type of AI model architecture behind many large language models) took off, so it calls itself a 'post-ChatGPT' company 1. * Its chip design can be shaped around large language model workloads, which may sidestep the memory capacity limits that have hit SRAM-only designs (chips that rely primarily on static random-access memory, a fast but typically lower-capacity type of memory) 1. * Situational Awareness invested in MatX. The firm was founded by former OpenAI researcher Leopold Aschenbrenner and it now manages more than $1.5 billion 2. * Aschenbrenner's fund argues that artificial general intelligence (AGI) is close, described as "in a couple of years" in Fortune's reporting. It targets the AI supply chain, including semiconductors, infrastructure, and power companies 2. ### A 2027 launch pits MatX against Nvidia's platform ecosystem, not just a chip * MatX plans to ship in 2027, so it will face NVIDIA's newer system platforms rather than today's hardware. * NVIDIA said Rubin-based products will be available from partners in the second half of 2026 3. * Beating a single chip on silicon area misses what matters, since NVIDIA wins through system-level integration. * NVIDIA pairs the GPU with its Vera CPU, NVLink 6 interconnect, and a mature software stack. That package frames Rubin as an integrated "AI factory" that startups struggle to match 4. * MatX may also need Taiwan Semiconductor Manufacturing Co.'s (TSMC's) scarce advanced packaging capacity, since advanced process and packaging capacity has become a strategic bottleneck in the industry 5. How would you feel if you could no longer use Tech in Asia? Share, tag us, and land on our Wall of!
MatX is Hiring for 42 Jobs on Simplify!
Find jobs on Simplify and start your career today
About MatX
Industries
Hardware
AI & Machine Learning
Company Size
51-200
Company Stage
Series B
Total Funding
$604.9M
Headquarters
Mountain View, California
Founded
2022
MatX is Hiring for 42 Jobs on Simplify!
Find jobs on Simplify and start your career today