[Engine] The vision encoder did not start #3645

Activity

1. github-actions

   added

   needs-formatTitle or template does not follow CONTRIBUTING.md yetTitle or template does not follow CONTRIBUTING.md yet

   on Oct 21, 2026
2. github-actions commented on Oct 21, 2026

   

   github-actionsBot

   on Oct 21, 2026 – with GitHub Actions

   More actions

   Thanks for the issue. To keep the tracker readable, every item needs an area prefix in the title and the parts of the issue form filled in. This one is missing:

   • Start the title with its area: [Engine], [GUI], [TUI], [Benchmark], [Docs] or [Other], then a short description. Example: "[Engine] decode stops after 2 minutes".

   Edit the issue (title and description) and this note goes away on its own. If it is still missing parts in 5 days it will be closed, which is not a rejection: fix it and comment /reopen.

   Details: CONTRIBUTING.md
3. Khalid-id

   changed the title [-]The vision encoder did not start[/-] [+][GUI] The vision encoder did not start[/+] on Oct 21, 2026
4. github-actions

   removed

   needs-formatTitle or template does not follow CONTRIBUTING.md yetTitle or template does not follow CONTRIBUTING.md yet

   on Oct 21, 2026
5. github-actions commented on Oct 21, 2026

   

   github-actionsBot

   on Oct 21, 2026 – with GitHub Actions

   More actions

   Thanks, this follows the format now.
6. Khalid-id

   changed the title [-][GUI] The vision encoder did not start[/-] [+][Engine] The vision encoder did not start[/+] on Oct 22, 2026
7. znbcr commented on Oct 22, 2026

   

   znbcr

   on Oct 22, 2026

   More actions

   The samething happend on mine ,and I have some test

   If I config the Conversation context more than 128k,and "Read pictures" is checked,then the engine start will be faile.

   My PC
   Operating System: Windows 10 Pro
   CPU: Intel 14900KF
   GPU: NVIDIA 2080ti (12 GB VRAM)
   RAM: 64 GB
8. Niko1221 commented on Oct 29, 2026

   

   Niko1221

   on Oct 29, 2026

   Owner

   More actions

   Please update to 0.2.7.1 and send the new error: it now names the encoder's last lines and exit status. Does it start with a context of 64K and Read pictures on, and does it fail only above 128K?
9. VSilantyev commented on Oct 30, 2026

   

   VSilantyev

   on Oct 30, 2026

   More actions

   Same issue on 0.2.8

   server_exited (exit code 1): RuntimeError: the vision encoder did not start: it ended without saying anything (killed by signal 6 (SIGABRT, Abort trap: 6); last lines of its log: dyld[5855]: Symbol not found: _posix_spawn_file_ac | During handling of the above exception, another exception occurred: | RuntimeError: the vision encoder did not start: it ended without saying anything (killed by signal 6 (SIGABRT, Abort trap: 6); last lines of its log: dyld[5858]: Symbol not found: _posix_spawn_file_ac

   Macbook Pro M1 Max 64gb
   Qwen 3.8
   MacOS 15.7.7 (24G720)
10. Khalid-id

    closed this as completedon Oct 31, 2026
11. Khalid-id commented on Oct 31, 2026

    

    Khalid-id

    on Oct 31, 2026

    Author

    More actions

    This issue is now fully resolved after updating to version 0.2.9. The model loads successfully with image reading enabled on both CPU and GPU, regardless of the context size (even above 128k).
    Thank you! I am closing this issue as fixed.