Skip to content

Model choice · project summaries · social tools

Choose a tag to compare

@yan-labs yan-labs released this 09 Jul 15:52

EN

Kollab release illustration

Model Selection

  • You can now choose from a clearer model lineup with Auto plus current Claude, GPT, Gemini, Grok, Kimi, DeepSeek, GLM, and Qwen-style routes.
  • Fixed model names, logos, descriptions, ordering, and message badges so the selector and task history describe the same model consistently.
  • Model credit pricing is more accurate for new routes, including GLM, GPT, Gemini, and discounted Ali Bailian calls.

Projects & Editor

  • Project summaries and recommended questions are now generated from project content instead of hardcoded fallback suggestions.
  • Task conversations can now insert an @mention when you preview a sidebar file, making it easier to ask about the exact document you opened.
  • Editor uploads now stay in the right conversation, project, or artifact scope, with cleaner attachment placement and fewer duplicate upload indicators.

Sharing & Social Tools

中文

Kollab release illustration

模型选择

  • 现在可以在更清晰的模型列表里选择 Auto,以及 Claude、GPT、Gemini、Grok、Kimi、DeepSeek、GLM 和通义类路线。
  • 修复了模型名称、Logo、描述、排序和消息徽章不一致的问题,选择器和任务历史现在会用同一套模型表达。
  • 新模型的 credit 计费更准确,覆盖 GLM、GPT、Gemini 以及阿里百炼折扣调用。

项目与编辑器

  • 项目摘要和推荐问题现在会基于项目内容生成,不再依赖固定兜底文案。
  • 在侧边栏预览文件时,任务对话可以直接插入 @mention,方便围绕当前打开的文件追问。
  • 编辑器上传现在会保持在正确的会话、项目或 Artifact 范围内,附件位置更清楚,也减少了重复上传提示。

分享与社交工具

日本語

Kollab release illustration

モデル選択

  • Auto に加えて、Claude、GPT、Gemini、Grok、Kimi、DeepSeek、GLM、Qwen 系ルートを含む、より分かりやすいモデル一覧から選べるようになりました。
  • モデル名、ロゴ、説明、並び順、メッセージバッジを修正し、セレクターとタスク履歴で同じモデル表記が使われるようにしました。
  • GLM、GPT、Gemini、Ali Bailian の割引ルートなど、新しいモデルルートの credit 計算がより正確になりました。

プロジェクトとエディター

  • プロジェクト要約とおすすめ質問は、固定の候補ではなくプロジェクト内容から生成されるようになりました。
  • サイドバーでファイルをプレビューしたとき、タスク会話に @mention を挿入できるようになり、開いている資料についてすぐ質問できます。
  • エディターのアップロードは、会話、プロジェクト、Artifact の正しいスコープに保持され、添付位置とアップロード状態が分かりやすくなりました。

共有とソーシャルツール

한국어

Kollab release illustration

모델 선택

  • 이제 Auto와 함께 Claude, GPT, Gemini, Grok, Kimi, DeepSeek, GLM, Qwen 계열 경로를 더 명확한 모델 목록에서 선택할 수 있습니다.
  • 모델 이름, 로고, 설명, 정렬, 메시지 배지를 정리해 선택 화면과 작업 기록이 같은 모델 표현을 사용합니다.
  • GLM, GPT, Gemini, Ali Bailian 할인 경로를 포함해 새 모델 라우트의 credit 계산이 더 정확해졌습니다.

프로젝트와 에디터

  • 프로젝트 요약과 추천 질문이 고정 문구가 아니라 프로젝트 콘텐츠를 바탕으로 생성됩니다.
  • 사이드바에서 파일을 미리 볼 때 작업 대화에 @mention을 넣을 수 있어, 방금 연 문서에 대해 바로 질문할 수 있습니다.
  • 에디터 업로드는 올바른 대화, 프로젝트, Artifact 범위에 유지되며 첨부 위치와 업로드 상태가 더 깔끔하게 표시됩니다.

공유와 소셜 도구