New default compiler setting
Now uses resources_used=0.25, letting you scale from 1 to 4 cores at runtime, no recompilation needed. Super handy for testing.
INNOVATION
-
New compiler default enables dynamic multi-core scaling without recompilation
By
–
-
LLM Upgrades: Windows Support and New Model Format
By
–
LLM upgrades
• Language models now supported on Windows
• New model format supports future AIPU versions [ Heads-up: Models from older SDK versions need recompiling or redownloading for future-proofing.] -
Metis 4-Chip PCIe Cards Enable Edge AI and LLM Deployments
By
–
New hardware support
Say hello to the new 4-chip Metis PCIe cards (16GB and 64GB).
Ideal for edge deployments needing serious compute: multi-stream apps, larger models, LLMs. -
Humanoid Robot Transforms Into Thanksgiving Chef and Butler
By
–
Humanoid robot transforms into a Thanksgiving chef and household butler 👩🍳
— Amitav Bhattacharjee (@bamitav) 26 novembre 2025
pic.twitter.com/6vr9nvvVFl#humanoidrobotics #humanoid #Robotics #Robot #AI@Eli_Krumova @RoryCrave @Fabriziobustama @bigfunduHumanoid robot transforms into a Thanksgiving chef and household butler #humanoidrobotics #humanoid #Robotics #Robot #AI @Eli_Krumova @RoryCrave @Fabriziobustama @bigfundu
-
YOLOv8 and YOLOv11 Computer Vision Models Released
By
–
New CV model highlights:
• Yolov8m-seg: instance segmentation
• Yolov8m-pose: keypoints
• Oriented bounding box detection (Yolo8 & Yolo11 variants)
• DenseNet-121, RegNet family for classification -
Voyager SDK now supports Ubuntu 24.04 and Python 3.12
By
–
Ubuntu 24.04 support
The community asked, and it’s here. No more Docker hacks; Voyager SDK now runs natively on 24.04. And yes, Python 3.12 too: better error messages, performance boosts, cleaner type hints. -

Voyager SDK v1.5 Released with Ubuntu 24.04 and Python 3.12
By
–
Voyager SDK v1.5 is out now, and it’s a big one.
Native Ubuntu 24.04, Python 3.12, new Metis PCIe support, fresh CV models, and upgraded LLM support. Let’s dig in… -
Claude.ai fixes context limit frustration with automatic compaction
By
–
In case you missed it, earlier this week we fixed one of the most common frustrations on http://
Claude.ai: hitting context limits mid-conversation. Claude now intelligently compacts earlier context automatically when you're nearing the limit so the chat can keep going. -
99% Performance Achievable with Smaller Language Models
By
–
i bet you could get 99% of the performance with less powerful LLMs
-

AI and Robotics Automating Scientific Discovery in Drug Development
By
–
The next breakthrough medicines & materials will emerge from code as much as the lab. While in China, I visited companies like XtalPi & Insilico Medicine. This is about automating science itself. It felt like a time warp, AI + robotics compressing years of work into weeks.