Deploy DeepSeek on your private enterprise cloud in < 24 hrs. ✅ DeepSeek R1 (600B, 70B, 32B).✅ Your private AWS/GCP cloud.✅ live in <24 hrs.✅ Scale to zero or up.✅ use cloud credits, commits.

We're excited to launch Lightning enterprise managed endpoints! Deploy @deepseek_ai (32B, 600B) on your enterprise @AWS or @googlecloud VPC in < 24 hrs.✅ DeepSeek R1 (32B, 600B).✅ Your private AWS/GCP account.✅ Scale to zero, or up on demand.✅ Live in <24 hrs.✅ Use cloud

RT @akshay_pachaar: I used @LightningAI ⚡️ Studio for developing this application!. The studio reads like a blog, encapsulating all my code….

Learn how the @deepseek_ai's R1 sausage is made. In our most recent Thunder Session, CTO @lantiga and Thomas Viehmann discuss the role of distilled models in the world of R1. Watch the full episode ➡️

Welcome to our live show!

The AI policy debate isn't just about regulation—it's about fostering open innovation. Amidst the buzz about @deepseek_ai's success our CEO, @_willfalcon, emphasizes the need for competition and open-source access to prevent tech giants from locking out smaller players. Read

Our CTO and @PyTorch Technical Advisory Council (TAC) Chair, @lantiga shares the 2025 priorities for the PyTorch TAC: .🎯 Develop an open, multi-cloud CI system.🎯 Enhance support for diverse AI accelerators.🎯 Cultivate a high-quality, user-centric ecosystem.🎯 Actively gather

Host your own private / secure @deepseek_ai using @LightningAI ⬇️.

RT @_willfalcon: Given the success of DeepSeek R1, just going to pull up my predictions from last year 😉. New predictions coming in a few….

RT @bhimrazyadav: 🚀 Finally, It's here! Chat with DeepSeek-R1 – Advanced AI Reasoning Model 🤖. 🔥 Try out the power of DeepSeek-R1 on @Light….

Get OpenAI-o1 level intelligence at 90% less cost with DeepSeek AI's DeepSeek-R1. This featured Studio by @Akshay guides you to build a fully self-hosted "Chat with your Docs" RAG application, served locally through @Ollama. Link to Studio ➡️

@Akshay @ollama oops! sorry accidentally tagged the wrong username @akshay_pachaar.

Get under the hood with @lantiga and Tom as they use traces and the PyTorch profiler to take a close look at what’s going on inside fp8 forward and backward. Watch the full Thunder Session below and tune in every Friday for the livestream. See episode 20 in full ➡️

Welcome to our live show!

Calling all London AI enthusiasts! Don’t miss our meetup on Jan 22nd. Join us for an evening of:.🔥 Inspiring talks from @aniketmaurya, @NoahSantoni, @ashvardanian, Evgeniia Grigoreva.🤝 Networking with AI researchers & developers.💡 Sharing work and demos.Save your spot now!

RT @_willfalcon: You're paying ~5x for GPUs without Lightning. GPU providers advertise cheap rates but lack persistent storage and CPU-to….

RT @mashaan_14: I posted a youtube video setting up a GPU VM on @LightningAI cloud to run nerfstudio. I also explained how to setup a web p….

It's the latest and greatest Thunder Sessions, where @lantiga & Tom discuss granular control and opening up possibilities for AI engineers and developers. See episode 19 in full here:

🔥 Introducing PyTorch Lightning 2.5.0!. New in this release:.✅ Support for Distributed Tensors and TorchAO for faster, more efficient training, both in PyTorch Lightning and Fabric. ✅ Stronger correctness guarantees when training is interrupted and then resumed, in all

RT @_willfalcon: Building an ML platform by stringing together ec2, logging, monitoring, autoscaling: etc sounds fun until this happens 😂….