The annual @hotchipsorg conference this week reminded us to share our article from the recent special issue of @ieeemicro, which highlights our Hot Chips 2024 presentation as one of the best at the event. 👀 Read the full “FuriosaAI RNGD: A Tensor Contraction Processor for

🔥 From our LG AI Research news to our $125 million funding raise, and from new executive hires to presenting six papers at ICML and ACL, it’s been a busy few months. We also managed to attend SuperAI in Singapore and RAISE Summit in Paris. Plus, we hosted our inaugural happy

Introducing Grok Imagine.

Last month, we presented four papers at ICML 2025 in Vancouver and two papers at ACL 2025 in Vienna. These six papers dig into ways to make advanced AI systems more efficient, more capable, and more flexible. We’re grateful to have collaborated on these projects with the

At OCP APAC last week and at OCP Korea this week, we presented “From Silicon to AI Serving: Optimizing Inference and Engineering What’s Next”, which covered everything from designing and manufacturing silicon to applying it to real-world AI services. Our Head of Product, Donggun

Thank you to everyone in the AI community we met in NYC this summer and those who came to our happy hour at @nyuniversity's Global AI Frontier Lab last week. It was a pleasure to show our AI accelerator, RNGD (pronounced “renegade”), to so many engineers and researchers working

The future of AI compute is brighter than ever. Seeing our AI accelerator, RNGD (pronounced “renegade”), in Times Square on the @Nasdaq billboard was a great way to mark our recent business momentum, including hosting a series of productive meetings at @nyuniversity's Global AI

Last week, we presented two papers at ACL in Vienna, authored by our engineers and researchers, with collaborators from @SeoulNatluni, @UWMadison, @berkeley_ai, @BerkeleyLab, and @UCBerkeley:. 1️⃣ State-offset Tuning: State-based Parameter-Efficient Fine-Tuning for State Space

If you’re in NYC on Friday, come listen to FuriosaAI co-founder and CEO June Paik and Professor of Computer Science and Data Science KyungHyun Cho speak at our inaugural Powered by Furiosa happy hour, moderated by Associate Professor and Furiosa Advisor Nuno Lopes:. 📅 August 8,

We’re thrilled to welcome Jeehoon Kang as our Chief Research Officer and Youngjin Cho as our Vice President of Hardware. Kang is a renowned expert in parallel systems from KAIST, with multiple Distinguished Paper Awards from top conferences like PLDI and POPL. He will lead the

After our LG AI Research and Series C bridge funding news, FuriosaAI CEO June Paik went on @CNBC to discuss the exciting momentum we’re seeing:. “Especially with the current fundraising and this strong LG reference, we are getting strong requests from the customer side to sample

We're proud to announce that we've raised $125M in Series C bridge funding to scale sustainable AI compute and accelerate our roadmap. This brings our total funding to $246 million and will help us meet growing demand from global enterprise customers for our flagship AI chip,

🛬 Next week, we’re hosting our inaugural happy hour: Powered by Furiosa. 🍹 On August 8, we’re bringing together industry leaders and practitioners at the forefront of AI to discuss, share knowledge, and enjoy great food and drinks. 🤝 If you work in AI research, engineering,

We are excited to announce that @LG_AI_Research has adopted our RNGD (pronounced “Renegade”) AI accelerator for inference computing with its EXAONE models. RNGD achieves 2.25x better LLM inference performance per watt vs. GPUs, while also meeting LG’s demanding latency and

1️⃣ Parameter-Efficient Fine-Tuning of State Space Models: 2️⃣ VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data: 3️⃣ TabFlex: Scaling Tabular Learning to Millions with Linear Attention:

This week, we presented four papers at @icmlconf in Vancouver, authored by our engineers and researchers, with collaborators from @UniversityKorea, @UCBerkeley, @UWMadison, @SeoulNatlUni, and @Microsoft:

“We use Rust. We assemble all the asynchronous parts in Rust, tying them to the hardware features, assemble the program, send it down, and run it. Because there’s effectively no runtime, the runtime overhead converges to almost 0. That’s why we can get good results in terms of

Thank you to everyone who stopped by our @RaiseSummit booth, watched our keynote, and drove the conversation on AI inference. 🤝 Couldn’t make it to #RaiseSummit or want to follow up? Reach out and let’s keep the conversation going:

We are excited to partner with turnkey AI cloud platform hosted·ai to deliver high-performance AI infrastructure built on our Tensor Contraction Processor (TCP) architecture. We expect hosted·ai’s cloud platform to support our flagship RNGD (pronounced “Renegade”) chip by the end

🛬 Next week, we’re at @RaiseSummit. 👀 Watch our co-founder and CEO, June Paik, keynote the Grace Hopper stage at 9:40 on July 9. 🤝 Connect with us in the #raisesummit app, stop by Booth 85 from 9:00 to 18:00 on July 8 and July 9, and book a meeting in advance:

📣 “It’s not healthy that there is just one architecture and just one direction,” our co-founder and CEO, June Paik, said at @superai_conf last month. “We call AI the age of discovery. I think on the compute side, we need the same approach. We need to explore new architectures.