LLMs are hungry beasts—serving them fast is expensive. But what if we could cut the cost without cutting corners?
Tilus, a new GPU virtual machine does just that—by unleashing low-precision math at any bit width, not just powers of 2.
SOFTWARE
-
Tilus GPU VM cuts LLM serving costs with low-precision math
By
–
-
Luma Labs Releases Ray2 Camera Concepts API for Developers
By
–
Luma Labs released Ray2 Camera Concepts in the API
— Rowan Cheung (@rowancheung) 29 avril 2025
The move allows developers to integrate the model’s advanced video controls (for cinematic framing and dynamic movement) into their applicationspic.twitter.com/1Ul8z0WmOaLuma Labs released Ray2 Camera Concepts in the API The move allows developers to integrate the model’s advanced video controls (for cinematic framing and dynamic movement) into their applications
-
TensorFlow.js Implementation with Canvas Training Visualization
By
–
"Write an implementation of this paper using Tensorflow.js and visualize the training with Canvas."
— Jeff Dean (@JeffDean) 28 avril 2025
Thanks, Gemini! https://t.co/718fKWV25V"Write an implementation of this paper using Tensorflow.js and visualize the training with Canvas." Thanks, Gemini!
-
Systems and Workflows for Efficiency Optimization
By
–
I'm all about systems and workflows for efficiency. That's what I actively seek out.
-
SIMT GPU Programming Model and Systolic Arrays Distinction
By
–
The SIMT programming model of GPUs doesn’t really have anything to do with systolic arrays, although the modern tensor cores lean a little closer. I860 pipeline mode had some more of the flavor for a single thread of floating point operations.
-
OpenSearch Evolves from Imitator to Innovator in Enterprise Search
By
–
OpenSearch has moved from imitator to innovator in enterprise search, and is increasingly getting lots of things right with its community. On the eve of @OpenSearchProj Con, I took a look at its last year of growth.
-
ChatGPT Now Available on WhatsApp for Live Answers
By
–
Search in WhatsApp You can now send a WhatsApp message to 1-800-ChatGPT (+1-800-242-8478) to get up-to-date answers and live sports scores. Accessible everywhere ChatGPT is available. https://
wa.me/18002428478 -
SE1 Development Continues Despite Earlier Pause
By
–
We didn’t forget about SE1, we are still developing it
-
Cursor Support Line Shared False Information With Users
By
–
I am looking for people who were affected by the recent problem with Cursor, where its support line shared untrue information with users (news.ycombinator.com/item?id…). If you were affected by this, please contact me at: cade.metz@nytimes.com
-
Game-Changing Product Updates: Lakeflow Connect and Genie APIs
By
–
We have a ton of recent product releases and updates that are game-changing for customers. Holly Smith and Nick Karpov dive into these updates and how they work together under one architecture. Hear their take on new Lakeflow Connect integrations, Genie Conversation APIs, Delta