What causes "Failed to delete images" in invoke-ai/InvokeAI?

Inspect server logs for the underlying exception

invoke-ai/InvokeAI · error · HTTPException

Failed to delete images

Error message

Failed to delete images

What it means

HTTP 500 raised by POST /images/delete when the bulk deletion service call throws an unexpected (non-HTTPException) error. Successful but partial deletions are reported in the DeleteImagesResult payload instead of raising, so this error means the operation crashed outright.

Source

Thrown at invokeai/app/api/routers/images.py:606

                # would answer for names the caller was never entitled to touch.
                #
                # This is narrow only because image_records.get() no longer translates a
                # sqlite3.Error into this exception — see the comment there. If that
                # translation ever comes back, a locked or corrupt database would land here
                # and a whole failed batch would answer 200 with empty result lists.
            except Exception:
                # A genuine deletion failure (not an auth/404 skip) — report it so the
                # client can surface a partial-failure warning, matching the video path.
                failed_images.add(image_name)
        return DeleteImagesResult(
            deleted_images=list(deleted_images),
            failed_images=list(failed_images),
            affected_boards=list(affected_boards),
        )
    except HTTPException:
        raise
    except Exception:
        raise HTTPException(status_code=500, detail="Failed to delete images")


@images_router.delete("/uncategorized", operation_id="delete_uncategorized_images", response_model=DeleteImagesResult)
def delete_uncategorized_images(
    current_user: CurrentUserOrDefault,
) -> DeleteImagesResult:
    """Deletes all uncategorized images owned by the current user (or all if admin)"""
    assert_image_move_maintenance_inactive()

    image_names = ApiDependencies.invoker.services.board_images.get_all_board_image_names_for_board(
        board_id="none", categories=None, is_intermediate=None
    )

    try:
        deleted_images: set[str] = set()
        failed_images: set[str] = set()
        affected_boards: set[str] = set()
        for image_name in image_names:

View on GitHub (pinned to 0b6a024f2f)

Solutions

Inspect server logs for the underlying exception
Retry the delete with a smaller batch of image_names
Verify image files and database are present and writable
Restore missing files or clean orphaned DB records before retrying

Defensive patterns

Strategy: try-catch

Validate before calling

// confirm images still exist before bulk delete
const existing = await api.getImageDtos(imageNames);
if (existing.length !== imageNames.length) {
  imageNames = existing.map(i => i.image_name);
}

Try / catch

try {
  const res = await api.deleteImagesFromList(imageNames);
  // inspect res.failed_images for partial failures
} catch (e) {
  if (e instanceof ApiError && e.status === 500) {
    // retry in smaller batches; check server logs
  }
}

Prevention

Delete in small batches to avoid timeouts
Never delete image files on disk manually while the app runs
Check failed_images in successful responses before assuming all deleted
Avoid deleting during active generation jobs

When it happens

Trigger: DELETE /api/v1/images/delete when the images service raises while deleting the listed image_names — e.g. storage layer failure, database error, or corrupted image records.

Common situations: Image files on disk already removed manually so record deletion fails; sqlite DB locked by another process; deleting many images triggers a storage backend timeout.

Related errors

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/0149966610c3f69f. Report an issue: GitHub.