💄 style: add wanxiang2.7 & keling ImageGen from Qwen by hezhijie0327 · Pull Request #13478 · lobehub/lobehub

hezhijie0327 · 2026-04-01T13:10:15Z

💻 Change Type

🔗 Related Issue

🔀 Description of Change

增加 wan2.7-image wan2.7-image-pro kling/kling-v3-image-generation kling/kling-v3-omni-image-generation
增加新的异步生图 Endpoint image-generation/generation 目前为新模型默认推荐 (Wan2.6, Wan2.6, Keling, etc)
增加 syncOnlyModels 模型列表,用于强制切换至同步生图 multimodal-generation/generation (qwen-image-edit/max, qwen-image-2.0, z-image-turbo)

Note:

Keling 模型需要到百炼平台开通后使用

模型	截图	日志
`kling/kling-v3-image-generation`
`kling/kling-v3-omni-image-generation`
`wan2.7-image`

🧪 How to Test

Tested locally
Added/updated tests
No tests needed

📸 Screenshots / Videos

Before	After
...	...

📝 Additional Information

vercel · 2026-04-01T13:10:22Z

@hezhijie0327 is attempting to deploy a commit to the LobeHub OSS Team on Vercel.

A member of the Team first needs to authorize it.

sourcery-ai

Sorry @hezhijie0327, you have reached your weekly rate limit of 500000 diff characters.

Please try again later or upgrade to continue using Sourcery

lobehubbot · 2026-04-01T13:11:19Z

@arvinxx Could you review this PR? It adds wanxiang2.7 and keling image generation models to the Qwen provider, including a new async image generation endpoint.

codecov · 2026-04-01T13:19:41Z

Codecov Report

❌ Patch coverage is 75.90361% with 40 lines in your changes missing coverage. Please review.
✅ Project coverage is 66.51%. Comparing base (aec2d30) to head (b72db3b).
⚠️ Report is 3 commits behind head on canary.

Additional details and impacted files

@@            Coverage Diff             @@
##           canary   #13478      +/-   ##
==========================================
+ Coverage   66.50%   66.51%   +0.01%     
==========================================
  Files        1995     1995              
  Lines      165151   165264     +113     
  Branches    15938    16726     +788     
==========================================
+ Hits       109834   109930      +96     
- Misses      55195    55212      +17     
  Partials      122      122

Flag	Coverage Δ
app	`58.30% <ø> (+0.01%)`	⬆️
database	`92.57% <ø> (ø)`
packages/agent-runtime	`88.98% <ø> (ø)`
packages/context-engine	`86.51% <ø> (ø)`
packages/conversation-flow	`92.36% <ø> (ø)`
packages/file-loaders	`87.02% <ø> (ø)`
packages/memory-user-memory	`66.68% <ø> (ø)`
packages/model-bank	`99.85% <ø> (ø)`
packages/model-runtime	`84.66% <75.90%> (-0.07%)`	⬇️
packages/prompts	`65.80% <ø> (ø)`
packages/python-interpreter	`92.90% <ø> (ø)`
packages/ssrf-safe-fetch	`0.00% <ø> (ø)`
packages/utils	`90.02% <ø> (ø)`
packages/web-crawler	`88.82% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
Store	`66.88% <ø> (ø)`
Services	`49.50% <ø> (ø)`
Server	`65.97% <ø> (+<0.01%)`	⬆️
Libs	`51.03% <ø> (ø)`
Utils	`91.01% <ø> (ø)`

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

sourcery-ai

Sorry @hezhijie0327, you have reached your weekly rate limit of 500000 diff characters.

Please try again later or upgrade to continue using Sourcery

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 9126fe0388

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

chatgpt-codex-connector · 2026-04-01T15:32:02Z

packages/model-runtime/src/providers/qwen/createImage.ts

  /^wan2\.(2|5)-t2i-/,
  /^wanx2\.(0|1)-t2i-/,
  /^wanx-v1/,
+  /^qwen-image(-plus)?$/,


Keep qwen-image models off legacy size-defaulted route

Adding ^qwen-image(-plus)?$ to text2ImageModels routes qwen-image and qwen-image-plus through createLegacySynthesisTask, which injects a default size: '1024*1024' whenever the caller omits size. These two models are declared with constrained size enums in packages/model-bank/src/aiModels/qwen.ts (1664x928, 1472x1140, 1328x1328, 1140x1472, 928x1664), so OpenAI-compatible calls without an explicit size can now send an unsupported size and fail at runtime. Before this commit they used the multimodal path, which did not force this default size.

Useful? React with 👍 / 👎.

@hezhijie0327 check 下这个

符合预期的, qwen-image 和 qwen-image-plus 支持 text2image 即 createLegacySynthesisTask

style: add wanxiang2.7 support style: add keling support feat: support new async method feat: support new async method feat: support new async method feat: support new async method

style: add wan2.7 videoGen models style: add wan2.7 videoGen models

vercel · 2026-04-03T08:04:50Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
lobehub	Ready	Preview, Comment	Apr 3, 2026 8:21am

tjx666 · 2026-04-03T08:25:04Z

Code review findings after verifying against full source code:

Issues

1. [high] wan2.6-image lost image-required validation — packages/model-runtime/src/providers/qwen/createImage.ts

imageRequiredModels includes /^wan2\.6-image/, but the validation check only exists inside createHTTPSyncGeneration (~line 236). Since the default route changed from sync to async, wan2.6-image now goes through createHTTPAsyncGenerationTask, which completely bypasses the image-required validation. Users calling wan2.6-image without providing an image will no longer get a helpful error message — they'll get a raw Alibaba API error instead.

Suggestion: move the imageRequiredModels validation up to the createQwenImage entry point, or duplicate it in createHTTPAsyncGenerationTask.

2. [medium] Default route changed from sync to async — behavioral change affects existing models

Previously, unmatched models defaulted to multimodal-generation/generation (sync). Now they default to image-generation/generation (async polling). Affected existing models include wan2.6-image and wan2.6-t2i. New models (wan2.7, kling) are added by this PR so they should be fine, but wan2.6-* models existed before — please confirm they work correctly on the new async image-generation endpoint.

3. [low] Error double-wrapping in createLegacySynthesisTask (pre-existing)

createLegacySynthesisTask throws AgentRuntimeError.createImage(...) directly for its missing-imageUrls validation (~line 102), but the outer createQwenImage catch wraps it again with AgentRuntimeError.createImage(...). This nests one AgentRuntimeError inside another, losing the inner errorType. By contrast, createHTTPAsyncGenerationTask throws plain Error which avoids this. Pre-existing issue, but could be cleaned up during this refactor.

4. [low] Redundant /r2v$/ in image2VideoModels — packages/model-runtime/src/providers/qwen/createVideo.ts

/r2v$/ appears in both image2VideoModels and reference2VideoModels. Due to the else if chain, reference2VideoModels (branch 5) is checked before image2VideoModels (branch 7), so r2v models always get reference-to-video treatment. The /r2v$/ entry in image2VideoModels is redundant and misleading — it could be removed without changing any behavior.

5. [low] Missing $ anchors on syncOnlyModels regexes

/^qwen-image-(edit|max)/, /^qwen-image-2\.0/, /^z-image-turbo/ all use prefix matching without $. This works for all current models, but future models with the same prefix but different API requirements would be silently misrouted. Consider adding a comment to clarify this is intentional prefix matching.

hezhijie0327 · 2026-04-03T08:59:19Z

FYI.

Fixed
Works as expected, approved
Fixed
Removed
Works as expected

This comment was translated by Claude.

Original Content

FYI.

已修复
符合预期,支持
已修复
已移除
符合预期

lobehubbot · 2026-04-03T09:17:38Z

❤️ Great PR @hezhijie0327 ❤️

The growth of project is inseparable from user feedback and contribution, thanks for your contribution! If you are interesting with the lobehub developer community, please join our discord and then dm @arvinxx or @canisminor1990. They will invite you to our private developer channel. We are talking about the lobe-chat development or sharing ai newsletter around the world.

@chriszf

# 🚀 release: 20260407 This release includes **148 commits**. Key updates are below. - **Response API tool execution is more capable and reliable** — Added hosted builtin tools + client-side function tools and improved tool-call streaming/completion behavior. [#13406](#13406) [#13414](#13414) [#13506](#13506) [#13555](#13555) - **Input and composition UX upgraded** — Added AI input auto-completion and multiple chat-input stability fixes. [#13458](#13458) [#13551](#13551) [#13481](#13481) - **Model/provider compatibility improved** — Better Gemini/Google tool schema handling and additional model updates. [#13429](#13429) [#13465](#13465) [#13613](#13613) - **Desktop and CLI reliability improved** — Gateway WebSocket support and desktop runtime upgrades. [#13608](#13608) [#13550](#13550) [#13557](#13557) - **Security hardening continued** — Fixed auth and sanitization risks and upgraded vulnerable dependencies. [#13535](#13535) [#13529](#13529) [#13479](#13479) ### Models & Providers - Added/updated support for `glm-5v-turbo`, GLM-5.1 updates, and qwen3.5-omni series. [#13487](#13487) [#13405](#13405) [#13422](#13422) - Added additional ImageGen providers/models (Wanxiang 2.7 and Keling from Qwen). [#13478](#13478) - Improved Gemini/Google tool schema and compatibility handling across runtime paths. [#13429](#13429) [#13465](#13465) [#13613](#13613) ### Response API & Runtime - Added hosted builtin tools in Response API and client-side function tool execution support. [#13406](#13406) [#13414](#13414) - Improved stream tool-call argument handling and `response.completed` output correctness. [#13506](#13506) [#13555](#13555) - Improved runtime error/context handling for intervention and provider edge cases. [#13420](#13420) [#13607](#13607) ### Desktop App - Bumped desktop dependencies and runtime integrations (`agent-browser`, `electron`). [#13550](#13550) [#13557](#13557) - Simplified desktop release channel setup by removing nightly release flow. [#13480](#13480) ### CLI - Added OpenClaw migration command. [#13566](#13566) - Added local device binding support for `lh agent run`. [#13277](#13277) - Added WebSocket gateway support and reconnect reliability improvements. [#13608](#13608) [#13418](#13418) ### Security - Removed risky `apiKey` fallback behavior in webapi auth path to prevent bypass risk. [#13535](#13535) - Sanitized HTML artifact rendering and iframe sandboxing to reduce XSS-to-RCE risk. [#13529](#13529) - Upgraded nodemailer to v8 to address SMTP command injection advisory. [#13479](#13479) ### Bug Fixes - Fixed image generation model default switch issues. [#13587](#13587) - Fixed subtopic re-fork message scope behavior and agent panel reset edge cases. [#13606](#13606) [#13556](#13556) - Fixed chat-input freeze on paste and mention plugin behavior. [#13551](#13551) [#13415](#13415) - Fixed auth/social sign-in and settings UX edge cases. [#13368](#13368) [#13392](#13392) [#13338](#13338) ### Credits Huge thanks to these contributors: @chriszf @hardy-one @Innei @lijian @neko @OctopusNote @rdmclin2 @rivertwilight @RylanCai @suyua9 @sxjeru @Tsuki @wangyk @WindSpiritSR @yizhuo @YuTengjing @hezhijie0327 @arvinxx

sourcery-ai bot reviewed Apr 1, 2026

View reviewed changes

hezhijie0327 force-pushed the wan2.7 branch from 4d815e2 to 9126fe0 Compare April 1, 2026 13:12

hezhijie0327 marked this pull request as ready for review April 1, 2026 15:27

sourcery-ai bot reviewed Apr 1, 2026

View reviewed changes

chatgpt-codex-connector bot reviewed Apr 1, 2026

View reviewed changes

tjx666 force-pushed the wan2.7 branch from 9126fe0 to d674e6f Compare April 2, 2026 03:46

hezhijie0327 force-pushed the wan2.7 branch 4 times, most recently from 5c19eae to 7ffee3d Compare April 2, 2026 09:39

tjx666 force-pushed the wan2.7 branch from 7ffee3d to b10190d Compare April 2, 2026 16:44

hezhijie0327 force-pushed the wan2.7 branch 2 times, most recently from 9ad2dcd to ef267ce Compare April 3, 2026 07:14

hezhijie0327 added 2 commits April 3, 2026 15:37

💄 style: add wanxiang2.7 & keling ImageGen from Qwen

3713ffe

style: add wanxiang2.7 support style: add keling support feat: support new async method feat: support new async method feat: support new async method feat: support new async method

style: add wan2.7 videoGen models

5afd530

style: add wan2.7 videoGen models style: add wan2.7 videoGen models

tjx666 force-pushed the wan2.7 branch from ef267ce to 5afd530 Compare April 3, 2026 07:37

vercel bot deployed to Preview April 3, 2026 08:21 View deployment

fix: try to fix suggestions

b72db3b

tjx666 merged commit 210f020 into lobehub:canary Apr 3, 2026
20 of 21 checks passed

hezhijie0327 deleted the wan2.7 branch April 3, 2026 09:18

nekomeowww mentioned this pull request Apr 7, 2026

🚀 release: 20260407 #13626

Merged

Uh oh!

Conversation

hezhijie0327 commented Apr 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💻 Change Type

🔗 Related Issue

🔀 Description of Change

🧪 How to Test

📸 Screenshots / Videos

📝 Additional Information

Uh oh!

vercel bot commented Apr 1, 2026

Uh oh!

sourcery-ai bot left a comment

Choose a reason for hiding this comment

Uh oh!

lobehubbot commented Apr 1, 2026

Uh oh!

codecov bot commented Apr 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

sourcery-ai bot left a comment

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector bot Apr 1, 2026

Choose a reason for hiding this comment

Uh oh!

arvinxx Apr 2, 2026

Choose a reason for hiding this comment

Uh oh!

hezhijie0327 Apr 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

vercel bot commented Apr 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

tjx666 commented Apr 3, 2026

Issues

Uh oh!

hezhijie0327 commented Apr 3, 2026 • edited by lobehubbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

lobehubbot commented Apr 3, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

hezhijie0327 commented Apr 1, 2026 •

edited

Loading

codecov bot commented Apr 1, 2026 •

edited

Loading

hezhijie0327 Apr 2, 2026 •

edited

Loading

vercel bot commented Apr 3, 2026 •

edited

Loading

hezhijie0327 commented Apr 3, 2026 •

edited by lobehubbot

Loading