Voyager 1.8 is out, and the best bits are buried

In case you haven’t bumped up to the latest Voyager yet, let’s take a look at the highlights. 😃 Of which there are plenty.

TL;DR: Axelera Server 150p (4-chip PCIe, 8GB) and Metis chip-down designs are supported, semantic segmentation is now a first-class task alongside face recognition and licence plates, Pipeline Builder gets a much more finalised API, and there are standalone Python model recipes you can copy and edit.

Three sleeper hits I think are actually the good stuff:

The RISC-V toolchain swap is a free speed-up. "LLVM 20 replaces GNU newlib" reads like housekeeping. It's actually a wheel going from ~200 MiB to ~20 MiB, about 900 MiB back on disk, and up to ~10% better inference on small, control-bound networks. Faster for doing nothing. Sounds like a Dire Straits parady song.

The Scheduler stops grabbing your card just because you looked at it. Connection is lazy now, and a pipeline with no op.load node never touches the device. Previously, merely referencing the scheduler claimed every AI core. If you've had two scripts fight over cores on one box, that's the fix.

Frame pacing saves a debugging session. ./inference.py yolov8n-coco video.mp4@30 paces a file like a live camera. That's the "perfect on the test video, falls over on the real camera" bug, now reproducible on your desk.

Before you upgrade, three that might bite:

• AxModel moves from v4 to v5 so existing models won't run and must be re-deployed or re-downloaded;
• the driver upgrade is mandatory (Linux metis-dkms 1.6.2, Windows MetisDriver-1.3.14);
• and Linux device nodes swapped colons for hyphens, so metis-0:1:0 is now metis-0-1-0. Symlinks cover the old form, but check any script that parses it. 👍

Check it out, and let me know when you’ve updated!

https://docs.axelera.ai/sdk/release-notes/