Skip to content

Images returned from Read tool don't work with OpenAI-compatible providers #11304

Closed
#11323
Closed
Images returned from Read tool don't work with OpenAI-compatible providers#11304
#11323
Assignees
rekram1-node
Labels
bugSomething isn't working

Description

@zhming0
zhming0
opened on Jan 30, 2026

Description

When using the Read tool to load an image file, the image is correctly displayed and analyzed when using Claude, but fails when using Kimi k2.5 or other OpenAI-compatible providers, even though these models support vision.

Pasting an image directly into OpenCode works with both Claude and Kimi k2.5. Only the Read tool path fails.

Plugins

None

OpenCode version

1.1.44

Steps to reproduce

  1. Configure OpenCode to use Kimi k2.5 (or another OpenAI-compatible vision model)
  2. Ask: "Use the Read tool to load /tmp/some-image.png and describe it"
  3. Observe that the image is not properly passed to the model

Screenshot and/or share link

N/A

Operating System

macOS

Terminal

ghostty

Activity

  1. zhming0
    added
    bugSomething isn't working
    on Jan 30, 2026
  2. github-actions
    assigned
    rekram1-node
    on Jan 30, 2026

  3. github-actions commented on Jan 30, 2026

    @github-actions
    github-actionsBot
    on Jan 30, 2026 – with GitHub Actions
    Contributor

    This issue might be a duplicate of existing issues. Please check:

    Feel free to ignore if none of these address your specific case.

  4. rekram1-node
    closed this as completedin #11323on Feb 5, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

bugSomething isn't working

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions