VeizikNVIDIA, Apple Silicon and edge builds are available now.Release notes →
veiziknative inference engine
For business Sign inDownload
Releases

Veizik for Apple Silicon — 2026.10.02-r6.1

A fix release over r6: the CUDA option-forwarding overflow r6 listed as a known issue is fixed, and `veizik keys` now names the file and the reason when it cannot create its lock instead of reporting that another keys command is running.

This page records the state as of 2026-10-05 and is not updated as the catalogue moves. For what runs now, see models.

Support

Models in this release

ModelTaskContextStatus
Qwen2.5-0.5BLanguage32,768Available
Qwen2.5-1.5BLanguage32,768Available
Qwen2.5-7BLanguage32,768Available
Qwen3-1.7BLanguage32,768Available
Qwen2.5-0.5B-InstructLanguage32,768Available
Qwen2.5-1.5B-InstructLanguage32,768Available
Qwen2.5-7B-InstructLanguage32,768Available
Qwen3-1.7B-BaseLanguage32,768Available
Platform
macOS on Apple Silicon (arm64)
Install size
16.3 MB, 124 files — no dependency to install first
Verification
The file name, size and SHA-256 are printed on this page and on the release page, and the two agree. The disk image is signed with a Developer ID and notarised, and carries SHA256SUMS covering its other 49 files, so `shasum -a 256 -c SHA256SUMS` checks the contents after you open it.
File
veizik-metal-2026.10.02-r6.1.dmg — no longer downloadable
Size
6,025,679 bytes
SHA-256
40c619a349f2fd2967b2bf6f7c9c8652a8b2009ea9810c1734039c22aca38ddc
Installation

Verify, install, check

zsh
$ shasum -a 256 -c SHA256SUMS
$ ./install.sh
$ veizik doctor

Full steps are in the documentation.

Fixed

Fixed issues

  • CUDA bundles: more than about 64 KB of options forwarded by veizik run to a CUDA bundle could overflow a buffer inside your own veizik process on Arm builds. It was never reachable from an Apple Silicon install, and x86_64 Linux was not affected.
  • veizik keys under a fresh home directory: it creates its cache folder, and when it cannot create its lock file it names the file and the reason rather than reporting that another keys command is running.
Known issues

Known issues in 2026.10.02-r6.1

  • /v1/embeddings answers 501 in this release.
  • The serve window is 4,096 tokens per request for every model in this release (prompt plus max_tokens); veizik run is not limited this way.
  • Qwen3-1.7B returns its reasoning text inside the answer; add /no_think to the prompt for a direct reply.
  • Offline use: the session key lasts 48 hours after an online start; restarting the Mac ends that window.
  • Not supported: Intel Macs, Windows, Linux. Headless activation is not included in this build.