Skip to content

fix: add Qwen3.5 and Qwen3.6 to vision model regex - #397

Open
wwsinagogogo wants to merge 2 commits into
CherryHQ:mainfrom
wwsinagogogo:main
Open

fix: add Qwen3.5 and Qwen3.6 to vision model regex#397
wwsinagogogo wants to merge 2 commits into
CherryHQ:mainfrom
wwsinagogogo:main

Conversation

@wwsinagogogo

Copy link
Copy Markdown

Summary

  • Add Qwen3.5 and Qwen3.6 series to vision model regex
  • Previous pattern qwen3-vl didn't match Qwen3.5/3.6 model IDs (e.g. Qwen3.6-35B) which don't contain 'vl' suffix
  • New pattern qwen3\.[5-9](?:-[\w-]+)? covers Qwen3.5 through Qwen3.9
  • Existing qwen3-vl pattern preserved

Test plan

  • Select Qwen3.5 or Qwen3.6 model in mobile app
  • Upload an image and send a message asking about the image
  • Verify the model receives the image and responds correctly

Titan and others added 2 commits May 26, 2026 09:07
Qwen3.5 and Qwen3.6 series are vision models but their IDs don't
contain 'vl' suffix (e.g. Qwen3.6-35B), so they were not matched
by the existing qwen3-vl pattern.

Co-Authored-by: Claude <claude-opus-4-6-20250514>
Replace generic error messages with detailed, localized messages for
each error type (service-not-allowed, no-speech, audio-capture, etc.).
On Android devices without a compatible speech service (e.g. MIUI without
Google Play Services), users now see a clear explanation instead of
the cryptic "Insufficient permissions" error.

Co-Authored-by: Claude <noreply@anthropic.com>
@chengcheng84

Copy link
Copy Markdown
Contributor

#398 不是一样的吗

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants