feat(chat): generate videos inline from the chat UI#55
Merged
Conversation
Video models are back in the chat model picker (tagged with a video badge); selecting one turns the input box into a text-to-video prompt. sendMessage routes to POST /v1/videos with a draft preset (384x224, 2s, 8 steps, ~2 min on M5 Max), polls job status with an inline progress bar, then embeds the finished mp4 as a playable video in the assistant bubble (fetched with auth into an object URL). Failures and queue/memory rejections surface as normal error messages. The text record (job id, wall time, seed) survives reloads; the blob URL does not, by design. --- 视频模型回到 chat 模型下拉 (带 video 角标); 选中后输入框即文生视频 入口. sendMessage 改路由到 POST /v1/videos (draft 档 384x224, 2 秒, 8 步, M5 Max 上约 2 分钟), 轮询进度条, 完成后在助手气泡内嵌可播放的 mp4 (带鉴权拉取转 object URL). 失败与排队/内存拒绝以普通错误消息呈现. 文字记录 (job id, 用时, seed) 可跨刷新留存, blob URL 刷新即失效 (设计内).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Answers the owner's ask: video generation directly in the built-in chat. Video models reappear in the picker with a video badge; sending a prompt with one selected submits a /v1/videos job (draft preset, ~2 min), shows an inline progress bar, and embeds the finished mp4 as a playable video in the bubble.
Implementation notes: pure frontend (chat.html + i18n en/zh/ru); uses only the /v1/videos API surface that was real-machine validated in #53; content is fetched with the auth header into an object URL (video tags cannot send Authorization). Text record (job id / wall time / seed) persists across reloads; the blob URL intentionally does not.
Test plan: i18n JSON validated; the API call sequence (create -> poll -> content) is byte-identical to the curl flow验证过的 E2E in #53; Alpine bindings follow the existing message-bubble patterns. Live verification on m5max after deploy.
Self-merge under the same in-session owner authorization as #53/#54 (continuation of the same feature mandate, "前后端都得完美实现").
响应 owner 需求: 聊天内直接生成视频. 视频模型带角标回到下拉; 选中后发送
即提交 /v1/videos job (draft 档约 2 分钟), 气泡内进度条 + 完成后内嵌播放.
纯前端改动; API 调用序列与 #53 真机验证的 E2E 完全一致. 部署后在 m5max
实机验证. 依 #53/#54 同一会话内 owner 授权自助合并.