v0.9.3
更新内容 / Updates
-
新增 ComfyUI 生图后端支持,可在管理界面配置 ComfyUI 服务地址、工作流和输出节点。
-
Added ComfyUI image-generation backend support with configurable service URL, workflow, and output nodes in the admin UI.
-
完善统一生图适配层,使 OpenAI Images 兼容后端与 ComfyUI 后端共享请求、结果和预览处理流程。
-
Improved the unified image-generation adapter so OpenAI Images-compatible and ComfyUI backends share request, result, and preview handling.
-
修复生图请求参数转换、结果交接、模型识别及管理界面的多个问题。
-
Fixed multiple issues in image request conversion, result handoff, model detection, and the admin UI.
-
增加并更新生图回归测试,覆盖 ComfyUI 适配和相关边界场景。
-
Added and updated image-generation regression tests covering the ComfyUI adapter and related edge cases.
验证 / Verification
- JavaScript syntax:
node --check app/web/static/app.jspassed. - Full test suite:
614 passed. - GitHub Actions test, package, release, and GHCR jobs completed successfully.
- Windows, Linux, and macOS ARM64 packages uploaded successfully.
Docker image:
docker pull ghcr.io/pricening/llm-aio-gateway:latest