Absolutely delighted to see FluidStack featured by a16z in this article about AI compute 😄 Fluidstack is the lowest-cost compute provider across all GPU clouds— proud of our team for making this happen! Huge thanks to Guido Appenzeller, Matt Bornstein, and martin_casado for

Useful comparison of the clouds for H/A100. Inaccurate info on Fluidstack though- to clarify, we offer the lowest prices. - H100 SXM w/ InfiniBand clusters at $1.75/hr ✅ - H100 PCIe on-demand at $1.99/hr ✅ - A100 80GB on-demand at $1.49/hr ✅ DM me for large H100 clusters

I've tried Google console, Amazon, Azure, but Fluidstack is definitely the best setup process I've seen so far. Great to get help from the cloud

Nous Research is excited to announce the closing of our $5.2 million seed financing round. We're proud to work with passionate, high-integrity partners that made this round possible, including co-leads Distributed Global and OSS Capiτal, with participation from Vipul Ved Prakash, founder

Antonio Lopez Teknium Belgrano mephisto Bowen Peng Shannon Sands Jai Burren ausboss huemin Riva Shivani We're especially grateful for the collaboration and contributions of Anjney Midha 🇺🇸 and the open source grant from a16z, to Fluidstack and Replicate and @togetherxyz for sponsoring our work via compute grants, as well as LDJ, @euclaise, qnguyen3, Wing Lian (caseus),

Introducing our new flagship LLM, Nous-Hermes 2 on Mixtral 8x7B. Our first model that was trained with RLHF, and the first model to beat Mixtral Instruct in the bulk of popular benchmarks! We are releasing the SFT only and SFT+DPO model, as well as a qlora adapter for the DPO

Announcing the new flagship 7B model in the Hermes series: Hermes 2-Mistral-7B DPO. A very special thanks to our compute sponsor for this run, Fluidstack. hf.co/NousResearch/N… This model was DPO'd from OpenHermes 2.5 and improved on all benchmarks tested - AGIEval,

Ambassador Paul Steinmetz met w/Stéphane Fisch (Argo Capital, focusing on new tech including AI) & César Maklary (FluidStack, UK, the largest non American independent AI cloud computing company), introducing both to Foreign Minister Xavier Bettel, at last Wed.LACC Business Award.

Since I got asked multiple times this week. Here is a fully self contained LLaMA2 built for ROCm/AMD from a Macbook with ZML. You can copy and unpack this tar on a machine _without_ ROCm installed and it will run on the GPU.

🚀2,500 GPUs deployed in just 48 hours! 🚀 Proud to partner with poolside as they redefine software development with AI. Just two days after signing, they had access to the speed, flexibility, and performance they needed to stay ahead.

Honoured to celebrate the UK’s AI ecosystem with the PM Keir Starmer, MP Peter Kyle and Matt Clifford today 🇬🇧 As a UK-born business, Fluidstack has supported the UK’s top AI companies for 7+ years. Today we work with top AI labs like Black Forest Labs, poolside, and UK’s

If completed as envisioned, the FluidStack project could tip the balance of AI development toward France and Europe. wsj.com/tech/ai/france… Gary Wu César Maklary Fluidstack Lfg 🚀

We are excited to announce Fluidstack's partnership with the French Ministry of the Economy to build a 1GW nuclear-powered AI cluster in France. We will deploy 500k+ GPUs to begin, and will continue to scale this cluster into one of the largest AI clusters on the planet.

César Maklary and Gary Wu were honored to be invited to a roundtable with President Emmanuel Macron, alongside Arthur Mensch (Mistral AI), Eiso Kant (poolside), Jamie Dimon (J.P. Morgan), Ahmed Yahia Al Idrissi (Mubadala), and 14 other handpicked leaders in AI to advise the

“If completed as envisioned, the Fluidstack project could tip the balance of AI development toward France and Europe.” 🚀 Fluidstack in the spotlight! 🚀 @Fluidstack was featured in The Wall Street Journal and Financial Times!

lego lithography means lego gpus as well

Today, AI can generate tons of code—but how do we know if it's good? That's why we built Sculptor: the first coding agent environment. Sculptor helps you catch issues, write tests, and improve your code—all while you work in your favorite editor.

To put the new 5GW AI campus in Abu Dhabi (UAE) into perspective. It would support up to 2.5 million NVIDIA B200s. That's bigger than all other major AI infrastructure announcements we've seen so far.

De retour de Paris — ravi de retrouver @Cesarmaklary Stéphane Fisch Arthur Mensch pour lancer les premiers préparatifs de #Nexus2050