FuriosaAI
@furiosa.ai
Our mission is to make AI computing sustainable, enabling everyone on Earth access to powerful AI.
AI is moving fast. From single-model prompts to agentic, multi-model systems running continuously in production, we’ve entered a new inference reality.
December 30, 2025 at 6:48 PM
AI is moving fast. From single-model prompts to agentic, multi-model systems running continuously in production, we’ve entered a new inference reality.
Exciting research news from FuriosaAI + KAIST!
Our collaborative project, RawMed, has been accepted at #NeurIPS2025. This pioneering framework synthesizes multi-table, time-series clinical data using text-based representations, compression techniques, and efficient autoregressive modeling.
Our collaborative project, RawMed, has been accepted at #NeurIPS2025. This pioneering framework synthesizes multi-table, time-series clinical data using text-based representations, compression techniques, and efficient autoregressive modeling.
December 3, 2025 at 8:13 PM
Exciting research news from FuriosaAI + KAIST!
Our collaborative project, RawMed, has been accepted at #NeurIPS2025. This pioneering framework synthesizes multi-table, time-series clinical data using text-based representations, compression techniques, and efficient autoregressive modeling.
Our collaborative project, RawMed, has been accepted at #NeurIPS2025. This pioneering framework synthesizes multi-table, time-series clinical data using text-based representations, compression techniques, and efficient autoregressive modeling.
We are honored to be named one of CRN's 10 Hottest Semiconductor Startups of 2025. This recognition highlights three recent critical milestones: our successful Series C funding, major customer validation from LG AI Research, and our first RNGD server product.
December 1, 2025 at 7:47 PM
We are honored to be named one of CRN's 10 Hottest Semiconductor Startups of 2025. This recognition highlights three recent critical milestones: our successful Series C funding, major customer validation from LG AI Research, and our first RNGD server product.
We're headed to #NeurIPS2025!
We’re excited to join the global AI community in San Diego next week to share research insights, connect with innovators, and explore the latest breakthroughs in machine learning.
📍 Stop by our booth S13 to meet the team and discover opportunities to collaborate.
We’re excited to join the global AI community in San Diego next week to share research insights, connect with innovators, and explore the latest breakthroughs in machine learning.
📍 Stop by our booth S13 to meet the team and discover opportunities to collaborate.
November 25, 2025 at 11:43 PM
We're headed to #NeurIPS2025!
We’re excited to join the global AI community in San Diego next week to share research insights, connect with innovators, and explore the latest breakthroughs in machine learning.
📍 Stop by our booth S13 to meet the team and discover opportunities to collaborate.
We’re excited to join the global AI community in San Diego next week to share research insights, connect with innovators, and explore the latest breakthroughs in machine learning.
📍 Stop by our booth S13 to meet the team and discover opportunities to collaborate.
🎉 What an incredible evening co-hosting The AI Frontier with Akamai!
It was inspiring to bring together leaders, experts, and researchers across industries to share real-world use cases, trends, and challenges in building scalable and sustainable AI.
It was inspiring to bring together leaders, experts, and researchers across industries to share real-world use cases, trends, and challenges in building scalable and sustainable AI.
November 21, 2025 at 9:27 PM
🎉 What an incredible evening co-hosting The AI Frontier with Akamai!
It was inspiring to bring together leaders, experts, and researchers across industries to share real-world use cases, trends, and challenges in building scalable and sustainable AI.
It was inspiring to bring together leaders, experts, and researchers across industries to share real-world use cases, trends, and challenges in building scalable and sustainable AI.
Join us for an evening of conversation and networking with leaders shaping the future of AI.
We are excited to co-host this in-person event with Akamai Technologies, bringing together leaders from across industries and borders to explore how to design and scale high-performance AI systems.
We are excited to co-host this in-person event with Akamai Technologies, bringing together leaders from across industries and borders to explore how to design and scale high-performance AI systems.
November 7, 2025 at 10:36 PM
Join us for an evening of conversation and networking with leaders shaping the future of AI.
We are excited to co-host this in-person event with Akamai Technologies, bringing together leaders from across industries and borders to explore how to design and scale high-performance AI systems.
We are excited to co-host this in-person event with Akamai Technologies, bringing together leaders from across industries and borders to explore how to design and scale high-performance AI systems.
The future of AI runs on efficient compute. At #APEC2025, we’re demonstrating what that looks like 🚀
Our RNGD chip was engineered from the ground up with AI-native architecture to redefine compute, delivering uncompromising performance with breakthrough power efficiency.
Our RNGD chip was engineered from the ground up with AI-native architecture to redefine compute, delivering uncompromising performance with breakthrough power efficiency.
October 31, 2025 at 6:31 AM
The future of AI runs on efficient compute. At #APEC2025, we’re demonstrating what that looks like 🚀
Our RNGD chip was engineered from the ground up with AI-native architecture to redefine compute, delivering uncompromising performance with breakthrough power efficiency.
Our RNGD chip was engineered from the ground up with AI-native architecture to redefine compute, delivering uncompromising performance with breakthrough power efficiency.
📌 We’re at #PyTorchCon this week!
As a global company, we are excited to see that 61 countries contributed to @pytorch.org this year, not to mention all the vLLM, DeepSpeed, and Ray updates.
🤝 Connect with us at Booth S13 on October 22 and October 23 to discuss your AI infrastructure stack.
As a global company, we are excited to see that 61 countries contributed to @pytorch.org this year, not to mention all the vLLM, DeepSpeed, and Ray updates.
🤝 Connect with us at Booth S13 on October 22 and October 23 to discuss your AI infrastructure stack.
October 22, 2025 at 6:48 PM
📌 We’re at #PyTorchCon this week!
As a global company, we are excited to see that 61 countries contributed to @pytorch.org this year, not to mention all the vLLM, DeepSpeed, and Ray updates.
🤝 Connect with us at Booth S13 on October 22 and October 23 to discuss your AI infrastructure stack.
As a global company, we are excited to see that 61 countries contributed to @pytorch.org this year, not to mention all the vLLM, DeepSpeed, and Ray updates.
🤝 Connect with us at Booth S13 on October 22 and October 23 to discuss your AI infrastructure stack.
Behind the scenes: Furiosa engineer Sanguk Park and CTO Hanjoon Kim share how our team served gpt-oss-120B at 5.8 ms TPOT using just two RNGD cards while maintaining exceptional power efficiency (under 180 W per card) for OpenAI's South Korean office grand opening.
October 17, 2025 at 6:05 PM
Behind the scenes: Furiosa engineer Sanguk Park and CTO Hanjoon Kim share how our team served gpt-oss-120B at 5.8 ms TPOT using just two RNGD cards while maintaining exceptional power efficiency (under 180 W per card) for OpenAI's South Korean office grand opening.
The 2025 OCP Global Summit is officially underway 🎉
Join FuriosaAI, hosted·ai, and BlueSky Compute for a happy hour. Connect with peers across the AI infrastructure, cloud, and systems communities over drinks and light bites. Meet the team and exchange insights on next-generation AI compute.
Join FuriosaAI, hosted·ai, and BlueSky Compute for a happy hour. Connect with peers across the AI infrastructure, cloud, and systems communities over drinks and light bites. Meet the team and exchange insights on next-generation AI compute.
October 15, 2025 at 3:57 PM
The 2025 OCP Global Summit is officially underway 🎉
Join FuriosaAI, hosted·ai, and BlueSky Compute for a happy hour. Connect with peers across the AI infrastructure, cloud, and systems communities over drinks and light bites. Meet the team and exchange insights on next-generation AI compute.
Join FuriosaAI, hosted·ai, and BlueSky Compute for a happy hour. Connect with peers across the AI infrastructure, cloud, and systems communities over drinks and light bites. Meet the team and exchange insights on next-generation AI compute.
Missed our latest SDK 2025.3.0 release? We’re making #furiousprogress unlocking new levels of performance and efficiency for large-scale models and agentic AI on RNGD.
📑 More details: furiosa.ai/blog/furiosa...
📋 Release notes: developer.furiosa.ai/latest/en/wh...
📑 More details: furiosa.ai/blog/furiosa...
📋 Release notes: developer.furiosa.ai/latest/en/wh...
October 10, 2025 at 7:10 PM
Missed our latest SDK 2025.3.0 release? We’re making #furiousprogress unlocking new levels of performance and efficiency for large-scale models and agentic AI on RNGD.
📑 More details: furiosa.ai/blog/furiosa...
📋 Release notes: developer.furiosa.ai/latest/en/wh...
📑 More details: furiosa.ai/blog/furiosa...
📋 Release notes: developer.furiosa.ai/latest/en/wh...
It’s been an exciting few days at #SDEC2025 ⚡
Thank you to everyone who visited our booth and joined our keynote and panel on Furiosa’s journey to building next-gen AI chips.
Drop by Booth 6307 today and tomorrow as the conference continues to meet the team and keep the conversation going.
Thank you to everyone who visited our booth and joined our keynote and panel on Furiosa’s journey to building next-gen AI chips.
Drop by Booth 6307 today and tomorrow as the conference continues to meet the team and keep the conversation going.
October 10, 2025 at 6:27 AM
It’s been an exciting few days at #SDEC2025 ⚡
Thank you to everyone who visited our booth and joined our keynote and panel on Furiosa’s journey to building next-gen AI chips.
Drop by Booth 6307 today and tomorrow as the conference continues to meet the team and keep the conversation going.
Thank you to everyone who visited our booth and joined our keynote and panel on Furiosa’s journey to building next-gen AI chips.
Drop by Booth 6307 today and tomorrow as the conference continues to meet the team and keep the conversation going.
We’re pleased to announce a new partnership between FuriosaAI and ByteBridge to advance next-generation AI infrastructure across the APAC region.
October 7, 2025 at 6:59 AM
We’re pleased to announce a new partnership between FuriosaAI and ByteBridge to advance next-generation AI infrastructure across the APAC region.
🛬 We’re headed to #SDEC2025 in Kuala Lumpur next week to explore the future of AI and semiconductors.
👀 Don’t miss our co-founder and CEO, June Paik, onstage October 10 as he shares how Furiosa is building next-gen AI chips for a more sustainable future.
👀 Don’t miss our co-founder and CEO, June Paik, onstage October 10 as he shares how Furiosa is building next-gen AI chips for a more sustainable future.
October 2, 2025 at 7:15 PM
🛬 We’re headed to #SDEC2025 in Kuala Lumpur next week to explore the future of AI and semiconductors.
👀 Don’t miss our co-founder and CEO, June Paik, onstage October 10 as he shares how Furiosa is building next-gen AI chips for a more sustainable future.
👀 Don’t miss our co-founder and CEO, June Paik, onstage October 10 as he shares how Furiosa is building next-gen AI chips for a more sustainable future.
Introducing Furiosa NXT RNGD Server 🚀
Engineered for efficient AI inference, the NXT RNGD Server hosts up to 8 RNGD accelerators, delivering:
⚡ 4 PFLOPS (FP8) compute
📦 384GB HBM + 12TB/s bandwidth
🔋 Just 3kW power draw (compared to 10.2kW for H100 SXM servers)
Engineered for efficient AI inference, the NXT RNGD Server hosts up to 8 RNGD accelerators, delivering:
⚡ 4 PFLOPS (FP8) compute
📦 384GB HBM + 12TB/s bandwidth
🔋 Just 3kW power draw (compared to 10.2kW for H100 SXM servers)
September 30, 2025 at 9:14 PM
Introducing Furiosa NXT RNGD Server 🚀
Engineered for efficient AI inference, the NXT RNGD Server hosts up to 8 RNGD accelerators, delivering:
⚡ 4 PFLOPS (FP8) compute
📦 384GB HBM + 12TB/s bandwidth
🔋 Just 3kW power draw (compared to 10.2kW for H100 SXM servers)
Engineered for efficient AI inference, the NXT RNGD Server hosts up to 8 RNGD accelerators, delivering:
⚡ 4 PFLOPS (FP8) compute
📦 384GB HBM + 12TB/s bandwidth
🔋 Just 3kW power draw (compared to 10.2kW for H100 SXM servers)
Last week, we demoed OpenAI’s open-weight gpt-oss 120B model running live on RNGD, our flagship AI accelerator.
Here’s a recording of the demo in action, showing that cutting-edge models can be deployed well within the existing power budgets of typical data centers:
Here’s a recording of the demo in action, showing that cutting-edge models can be deployed well within the existing power budgets of typical data centers:
September 17, 2025 at 2:06 PM
Last week, we demoed OpenAI’s open-weight gpt-oss 120B model running live on RNGD, our flagship AI accelerator.
Here’s a recording of the demo in action, showing that cutting-edge models can be deployed well within the existing power budgets of typical data centers:
Here’s a recording of the demo in action, showing that cutting-edge models can be deployed well within the existing power budgets of typical data centers:
Today we partnered with OpenAI for the grand opening of its new Seoul office, where we showcased its new open-source gpt-oss 120B model running live on RNGD.
We demonstrated a real-time chatbot efficiently running the model on just two RNGD cards, using MXFP4 precision: furiosa.ai/blog/furiosa...
We demonstrated a real-time chatbot efficiently running the model on just two RNGD cards, using MXFP4 precision: furiosa.ai/blog/furiosa...
September 11, 2025 at 2:30 PM
Today we partnered with OpenAI for the grand opening of its new Seoul office, where we showcased its new open-source gpt-oss 120B model running live on RNGD.
We demonstrated a real-time chatbot efficiently running the model on just two RNGD cards, using MXFP4 precision: furiosa.ai/blog/furiosa...
We demonstrated a real-time chatbot efficiently running the model on just two RNGD cards, using MXFP4 precision: furiosa.ai/blog/furiosa...
Day 1 at #AIInfraSummit was energizing. The conversations at our booth make it clear: the future of AI depends not just on smarter models, but on more efficient compute.
That’s why we were excited to have Alex Liu onstage sharing how we are redefining performance and efficiency in AI infrastructure.
That’s why we were excited to have Alex Liu onstage sharing how we are redefining performance and efficiency in AI infrastructure.
September 10, 2025 at 2:24 PM
Day 1 at #AIInfraSummit was energizing. The conversations at our booth make it clear: the future of AI depends not just on smarter models, but on more efficient compute.
That’s why we were excited to have Alex Liu onstage sharing how we are redefining performance and efficiency in AI infrastructure.
That’s why we were excited to have Alex Liu onstage sharing how we are redefining performance and efficiency in AI infrastructure.
🛬 Next week, we’re at #AIInfraSummit.
👀 Watch our SVP of Product and Business, Alex Liu, present on the Enterprise AI stage at 4:10 PM on September 9.
🤝 Connect with us in the app, stop by Booth 728 from September 9 to September 11, and book a meeting in advance: lp.furiosa.ai/ai-infra-sum...
👀 Watch our SVP of Product and Business, Alex Liu, present on the Enterprise AI stage at 4:10 PM on September 9.
🤝 Connect with us in the app, stop by Booth 728 from September 9 to September 11, and book a meeting in advance: lp.furiosa.ai/ai-infra-sum...
September 5, 2025 at 2:13 PM
🛬 Next week, we’re at #AIInfraSummit.
👀 Watch our SVP of Product and Business, Alex Liu, present on the Enterprise AI stage at 4:10 PM on September 9.
🤝 Connect with us in the app, stop by Booth 728 from September 9 to September 11, and book a meeting in advance: lp.furiosa.ai/ai-infra-sum...
👀 Watch our SVP of Product and Business, Alex Liu, present on the Enterprise AI stage at 4:10 PM on September 9.
🤝 Connect with us in the app, stop by Booth 728 from September 9 to September 11, and book a meeting in advance: lp.furiosa.ai/ai-infra-sum...
Our engineers tackle groundbreaking challenges head-on. This month, we feature Seung Ho Song, a key engineer on our compiler team, and ask him about his experience.
Read the full spotlight in our newsletter: www.linkedin.com/pulse/furios...
Read the full spotlight in our newsletter: www.linkedin.com/pulse/furios...
September 3, 2025 at 2:31 PM
Our engineers tackle groundbreaking challenges head-on. This month, we feature Seung Ho Song, a key engineer on our compiler team, and ask him about his experience.
Read the full spotlight in our newsletter: www.linkedin.com/pulse/furios...
Read the full spotlight in our newsletter: www.linkedin.com/pulse/furios...
The annual Hot Chips conference this week reminded us to share our article from the recent special issue of IEEE Micro, which highlights our Hot Chips 2024 presentation as one of the best at the event.
👀 Read the full article here: dxttx52ei7rol.cloudfront.net/Micro_202503...
👀 Read the full article here: dxttx52ei7rol.cloudfront.net/Micro_202503...
August 27, 2025 at 1:45 PM
The annual Hot Chips conference this week reminded us to share our article from the recent special issue of IEEE Micro, which highlights our Hot Chips 2024 presentation as one of the best at the event.
👀 Read the full article here: dxttx52ei7rol.cloudfront.net/Micro_202503...
👀 Read the full article here: dxttx52ei7rol.cloudfront.net/Micro_202503...
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.
These six papers dig into ways to make advanced AI systems more efficient, more capable, and more flexible.
August 19, 2025 at 2:35 PM
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.
These six papers dig into ways to make advanced AI systems more efficient, more capable, and more flexible.
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.
August 15, 2025 at 2:49 PM
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.
Thank you to everyone in the AI community we met in NYC this summer and those who came to our happy hour at @nyu.edu’s Global AI Frontier Lab last week.
August 14, 2025 at 1:56 PM
Thank you to everyone in the AI community we met in NYC this summer and those who came to our happy hour at @nyu.edu’s Global AI Frontier Lab last week.
Last week, we presented two papers at ACL in Vienna, authored by our engineers and researchers, with collaborators from Seoul National University, @uwmadison.bsky.social, Berkeley Artificial Intelligence Research, @berkeleylab.lbl.gov, and @ucberkeleyofficial.bsky.social:
August 7, 2025 at 3:22 PM
Last week, we presented two papers at ACL in Vienna, authored by our engineers and researchers, with collaborators from Seoul National University, @uwmadison.bsky.social, Berkeley Artificial Intelligence Research, @berkeleylab.lbl.gov, and @ucberkeleyofficial.bsky.social: