From 1aef8951bedbb2f6e85da67babb67a08a271091c Mon Sep 17 00:00:00 2001 From: "Daxiong (Lin)" Date: Thu, 19 Nov 2026 12:48:49 +0800 Subject: [PATCH] changelog: add v0.45.5, v0.45.6 and v0.46.0 release notes Docs blocks for the two 0.45.x patch tags and v0.46.0, plus the CMS staging for comfyui (three versions) and cloud (v0.45.6, the version Cloud pins). Registry records the 22 published entries. --- .github/scripts/cms/published-versions.json | 56 +++++++++++++++++++ .../cms/staging/cloud/en/changelog/index.mdx | 9 +++ .../cms/staging/cloud/es/changelog/index.mdx | 9 +++ .../cms/staging/cloud/fr/changelog/index.mdx | 9 +++ .../cms/staging/cloud/ja/changelog/index.mdx | 9 +++ .../cms/staging/cloud/ko/changelog/index.mdx | 9 +++ .../cms/staging/cloud/ru/changelog/index.mdx | 9 +++ .../cms/staging/cloud/zh/changelog/index.mdx | 9 +++ .../cms/staging/en/changelog/index.mdx | 34 +++++++++++ .../cms/staging/es/changelog/index.mdx | 34 +++++++++++ .../cms/staging/fr/changelog/index.mdx | 34 +++++++++++ .../cms/staging/ja/changelog/index.mdx | 34 +++++++++++ .../cms/staging/ko/changelog/index.mdx | 34 +++++++++++ .../cms/staging/ru/changelog/index.mdx | 34 +++++++++++ .../cms/staging/zh/changelog/index.mdx | 34 +++++++++++ changelog/index.mdx | 49 ++++++++++++++++ es/changelog/index.mdx | 54 +++++++++++++++++- ja/changelog/index.mdx | 54 +++++++++++++++++- ko/changelog/index.mdx | 54 +++++++++++++++++- zh/changelog/index.mdx | 54 +++++++++++++++++- 20 files changed, 618 insertions(+), 4 deletions(-) diff --git a/.github/scripts/cms/published-versions.json b/.github/scripts/cms/published-versions.json index 0ff319ca5c..64d4d69796 100644 --- a/.github/scripts/cms/published-versions.json +++ b/.github/scripts/cms/published-versions.json @@ -1561,6 +1561,62 @@ "zh" ], "published_at": "2026-11-16T13:41:09.182Z" + }, + { + "project": "comfyui", + "version": "0.45.5", + "locales": [ + "en", + "es", + "fr", + "ja", + "ko", + "ru", + "zh" + ], + "published_at": "2026-11-19T04:46:13.606Z" + }, + { + "project": "cloud", + "version": "0.45.6", + "locales": [ + "en", + "es", + "fr", + "ja", + "ko", + "ru", + "zh" + ], + "published_at": "2026-11-19T04:46:56.119Z" + }, + { + "project": "comfyui", + "version": "0.45.6", + "locales": [ + "en", + "es", + "fr", + "ja", + "ko", + "ru", + "zh" + ], + "published_at": "2026-11-19T04:46:22.121Z" + }, + { + "project": "comfyui", + "version": "0.46.0", + "locales": [ + "en", + "es", + "fr", + "ja", + "ko", + "ru", + "zh" + ], + "published_at": "2026-11-19T04:46:30.034Z" } ] } diff --git a/.github/scripts/cms/staging/cloud/en/changelog/index.mdx b/.github/scripts/cms/staging/cloud/en/changelog/index.mdx index f6b723d971..66e6fca86c 100644 --- a/.github/scripts/cms/staging/cloud/en/changelog/index.mdx +++ b/.github/scripts/cms/staging/cloud/en/changelog/index.mdx @@ -4,6 +4,15 @@ cmsStaging: true --- + + +**Partner Node Updates** +* [**Kling 4.0 Image**](https://links.comfy.org/4j3lZ6E): Added support for the Kling 4.0 Image and Long Image +* [**HeyGen Avatar 6**](https://links.comfy.org/4kYSHZx): Added support for the HeyGen Avatar 6 +* [**Bria RMBG 3.0**](https://links.comfy.org/4j3ma1O): Added RMBG 3.0 to the Bria Remove Image Background node + + + **Partner Node Updates** diff --git a/.github/scripts/cms/staging/cloud/es/changelog/index.mdx b/.github/scripts/cms/staging/cloud/es/changelog/index.mdx index 6dd58a3b66..46d37ed442 100644 --- a/.github/scripts/cms/staging/cloud/es/changelog/index.mdx +++ b/.github/scripts/cms/staging/cloud/es/changelog/index.mdx @@ -4,6 +4,15 @@ cmsStaging: true --- + + +**Actualizaciones de nodos de socios** +* [**Kling 4.0 Image**](https://links.comfy.org/4j3lZ6E): Se agregó soporte para Kling 4.0 Image y Long Image +* [**HeyGen Avatar 6**](https://links.comfy.org/4kYSHZx): Se agregó soporte para HeyGen Avatar 6 +* [**Bria RMBG 3.0**](https://links.comfy.org/4j3ma1O): Se agregó RMBG 3.0 al nodo Bria: Quitar fondo de imagen + + + **Actualizaciones de nodos de socios** diff --git a/.github/scripts/cms/staging/cloud/fr/changelog/index.mdx b/.github/scripts/cms/staging/cloud/fr/changelog/index.mdx index 3d1446047f..a0796ee0e5 100644 --- a/.github/scripts/cms/staging/cloud/fr/changelog/index.mdx +++ b/.github/scripts/cms/staging/cloud/fr/changelog/index.mdx @@ -4,6 +4,15 @@ cmsStaging: true --- + + +**Mises à jour des nœuds partenaires** +* [**Kling 4.0 Image**](https://links.comfy.org/4j3lZ6E) : Ajout de la prise en charge de Kling 4.0 Image et Long Image +* [**HeyGen Avatar 6**](https://links.comfy.org/4kYSHZx) : Ajout de la prise en charge de HeyGen Avatar 6 +* [**Bria RMBG 3.0**](https://links.comfy.org/4j3ma1O) : Ajout de RMBG 3.0 au nœud Bria Remove Image Background + + + **Mises à jour des nœuds partenaires** diff --git a/.github/scripts/cms/staging/cloud/ja/changelog/index.mdx b/.github/scripts/cms/staging/cloud/ja/changelog/index.mdx index 97c795ebe3..ab48d54fae 100644 --- a/.github/scripts/cms/staging/cloud/ja/changelog/index.mdx +++ b/.github/scripts/cms/staging/cloud/ja/changelog/index.mdx @@ -4,6 +4,15 @@ cmsStaging: true --- + + +**パートナーノードの更新** +* [**Kling 4.0 Image**](https://links.comfy.org/4j3lZ6E): Kling 4.0 Image と Long Image のサポートを追加 +* [**HeyGen Avatar 6**](https://links.comfy.org/4kYSHZx): HeyGen Avatar 6 のサポートを追加 +* [**Bria RMBG 3.0**](https://links.comfy.org/4j3ma1O): Bria 画像背景除去ノードに RMBG 3.0 を追加 + + + **パートナーノードの更新** diff --git a/.github/scripts/cms/staging/cloud/ko/changelog/index.mdx b/.github/scripts/cms/staging/cloud/ko/changelog/index.mdx index 57998cc40c..ff67a94346 100644 --- a/.github/scripts/cms/staging/cloud/ko/changelog/index.mdx +++ b/.github/scripts/cms/staging/cloud/ko/changelog/index.mdx @@ -4,6 +4,15 @@ cmsStaging: true --- + + +**파트너 노드 업데이트** +* [**Kling 4.0 Image**](https://links.comfy.org/4j3lZ6E): Kling 4.0 Image 및 Long Image 지원이 추가되었습니다 +* [**HeyGen Avatar 6**](https://links.comfy.org/4kYSHZx): HeyGen Avatar 6 지원이 추가되었습니다 +* [**Bria RMBG 3.0**](https://links.comfy.org/4j3ma1O): Bria 이미지 배경 제거 노드에 RMBG 3.0이 추가되었습니다 + + + **파트너 노드 업데이트** diff --git a/.github/scripts/cms/staging/cloud/ru/changelog/index.mdx b/.github/scripts/cms/staging/cloud/ru/changelog/index.mdx index f734d069ae..0c84770138 100644 --- a/.github/scripts/cms/staging/cloud/ru/changelog/index.mdx +++ b/.github/scripts/cms/staging/cloud/ru/changelog/index.mdx @@ -4,6 +4,15 @@ cmsStaging: true --- + + +**Обновления партнёрских узлов** +* [**Kling 4.0 Image**](https://links.comfy.org/4j3lZ6E): Добавлена поддержка Kling 4.0 Image и Long Image +* [**HeyGen Avatar 6**](https://links.comfy.org/4kYSHZx): Добавлена поддержка HeyGen Avatar 6 +* [**Bria RMBG 3.0**](https://links.comfy.org/4j3ma1O): RMBG 3.0 добавлена в узел Bria Remove Image Background + + + **Обновления партнёрских нод** diff --git a/.github/scripts/cms/staging/cloud/zh/changelog/index.mdx b/.github/scripts/cms/staging/cloud/zh/changelog/index.mdx index 874850369c..ba0af06e69 100644 --- a/.github/scripts/cms/staging/cloud/zh/changelog/index.mdx +++ b/.github/scripts/cms/staging/cloud/zh/changelog/index.mdx @@ -4,6 +4,15 @@ cmsStaging: true --- + + +**合作伙伴节点更新** +* [**Kling 4.0 Image**](https://links.comfy.org/4j3lZ6E):新增对 Kling 4.0 Image 和 Long Image 的支持 +* [**HeyGen Avatar 6**](https://links.comfy.org/4kYSHZx):新增对 HeyGen Avatar 6 的支持 +* [**Bria RMBG 3.0**](https://links.comfy.org/4j3ma1O):在 Bria 移除图像背景节点中新增 RMBG 3.0 + + + **合作伙伴节点更新** diff --git a/.github/scripts/cms/staging/en/changelog/index.mdx b/.github/scripts/cms/staging/en/changelog/index.mdx index 69ba489c65..06097038a0 100644 --- a/.github/scripts/cms/staging/en/changelog/index.mdx +++ b/.github/scripts/cms/staging/en/changelog/index.mdx @@ -4,6 +4,40 @@ cmsStaging: true --- + + +**New Open-Source Model Support** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): Video generation model support in core +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): Thinking, interleaved generation and self-correction, adds the ModelSamplingSenseNova node +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): Supported in the Generate Text node + +**Partner Node Updates** +* [**Kling 4.0 aspect ratios**](https://github.com/Comfy-Org/ComfyUI/pull/17815): 4:3 and 3:4 added to the Kling 4.0 nodes +* [**Beeble and Higgsfield on Comfy Router**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchX and Genjutsu motion transfer moved to the Router backend +* [**Fish Audio speech-to-text**](https://github.com/Comfy-Org/ComfyUI/pull/17828): Now runs transcribe-1-pro +* [**Claude Sonnet 4.5 removed**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Retired from the Anthropic Claude node + +**New Nodes** +* [**Video Deflicker**](https://github.com/Comfy-Org/ComfyUI/pull/17754): Removes flicker from video, with a faster RAFT optical flow +* [**Split Image Channels**](https://github.com/Comfy-Org/ComfyUI/pull/13548): Red, green, blue and alpha outputs from one image + + + + + +**Partner Node Updates** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): Image generation and editing with up to 10 reference images, long layouts +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): Reference to Video and Identity to Video at 1080p and 2K + + + + + +**Partner Node Updates** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): New model option on the Bria Remove Image Background node + + + **Partner Node Updates** diff --git a/.github/scripts/cms/staging/es/changelog/index.mdx b/.github/scripts/cms/staging/es/changelog/index.mdx index 1d3f800044..144a52b0e1 100644 --- a/.github/scripts/cms/staging/es/changelog/index.mdx +++ b/.github/scripts/cms/staging/es/changelog/index.mdx @@ -4,6 +4,40 @@ cmsStaging: true --- + + +**Nuevo soporte para modelos de código abierto** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): Soporte para el modelo de generación de video en el núcleo +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): Pensamiento, generación intercalada y autocorrección; añade el nodo ModelSamplingSenseNova +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): Compatible con el nodo Generar texto + +**Actualizaciones de nodos de socios** +* [**Relaciones de aspecto de Kling 4.0**](https://github.com/Comfy-Org/ComfyUI/pull/17815): Se añaden 4:3 y 3:4 a los nodos Kling 4.0 +* [**Beeble y Higgsfield en Comfy Router**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchX y la transferencia de movimiento Genjutsu se trasladaron al backend Router +* [**Fish Audio de voz a texto**](https://github.com/Comfy-Org/ComfyUI/pull/17828): Ahora ejecuta transcribe-1-pro +* [**Claude Sonnet 4.5 eliminado**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Retirado del nodo Anthropic Claude + +**Nuevos nodos** +* [**Eliminación de parpadeo de video**](https://github.com/Comfy-Org/ComfyUI/pull/17754): Elimina el parpadeo del video, con un flujo óptico RAFT más rápido +* [**Dividir canales de imagen**](https://github.com/Comfy-Org/ComfyUI/pull/13548): Salidas de rojo, verde, azul y alfa desde una sola imagen + + + + + +**Actualizaciones de nodos de socios** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): Generación y edición de imágenes con hasta 10 imágenes de referencia, diseños largos +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): Referencia a video e Identidad a video en 1080p y 2K + + + + + +**Actualizaciones de nodos de socios** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): Nueva opción de modelo en el nodo Bria: Quitar fondo de imagen + + + **Actualizaciones de nodos de socios** diff --git a/.github/scripts/cms/staging/fr/changelog/index.mdx b/.github/scripts/cms/staging/fr/changelog/index.mdx index f1a9835069..adc7cfa672 100644 --- a/.github/scripts/cms/staging/fr/changelog/index.mdx +++ b/.github/scripts/cms/staging/fr/changelog/index.mdx @@ -4,6 +4,40 @@ cmsStaging: true --- + + +**Prise en charge de nouveaux modèles open source** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755) : prise en charge du modèle de génération vidéo dans le cœur +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032) : raisonnement, génération entrelacée et autocorrection, ajoute le nœud ModelSamplingSenseNova +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787) : pris en charge dans le nœud Generate Text + +**Mises à jour des nœuds partenaires** +* [**Ratios d’aspect de Kling 4.0**](https://github.com/Comfy-Org/ComfyUI/pull/17815) : 4:3 et 3:4 ajoutés aux nœuds Kling 4.0 +* [**Beeble et Higgsfield sur Comfy Router**](https://github.com/Comfy-Org/ComfyUI/pull/17794) : le transfert de mouvement SwitchX et Genjutsu est déplacé vers le backend Router +* [**Transcription audio en texte de Fish Audio**](https://github.com/Comfy-Org/ComfyUI/pull/17828) : exécute désormais transcribe-1-pro +* [**Claude Sonnet 4.5 supprimé**](https://github.com/Comfy-Org/ComfyUI/pull/17819) : retiré du nœud Anthropic Claude + +**Nouveaux nœuds** +* [**Video Deflicker**](https://github.com/Comfy-Org/ComfyUI/pull/17754) : supprime le scintillement des vidéos, avec un flux optique RAFT plus rapide +* [**Split Image Channels**](https://github.com/Comfy-Org/ComfyUI/pull/13548) : sorties rouge, vert, bleu et alpha à partir d’une seule image + + + + + +**Mises à jour des nœuds partenaires** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797) : Génération et édition d'images avec jusqu'à 10 images de référence, dispositions longues +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793) : Reference to Video et Identity to Video en 1080p et 2K + + + + + +**Mises à jour des nœuds partenaires** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780) : Nouvelle option de modèle sur le nœud Bria Remove Image Background + + + **Mises à jour des nœuds partenaires** diff --git a/.github/scripts/cms/staging/ja/changelog/index.mdx b/.github/scripts/cms/staging/ja/changelog/index.mdx index 9b954d54f1..5aa76b1834 100644 --- a/.github/scripts/cms/staging/ja/changelog/index.mdx +++ b/.github/scripts/cms/staging/ja/changelog/index.mdx @@ -4,6 +4,40 @@ cmsStaging: true --- + + +**新しいオープンソースモデルのサポート** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): コアでビデオ生成モデルをサポート +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): 思考、インターリーブ生成、自己修正に対応し、ModelSamplingSenseNovaノードを追加 +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): TextGenerateノードでサポート + +**パートナーノードの更新** +* [**Kling 4.0のアスペクト比**](https://github.com/Comfy-Org/ComfyUI/pull/17815): Kling 4.0ノードに4:3と3:4を追加 +* [**Comfy Router上のBeebleとHiggsfield**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchXとGenjutsuのモーショントランスファーをRouterバックエンドに移行 +* [**Fish Audioの音声テキスト変換**](https://github.com/Comfy-Org/ComfyUI/pull/17828): transcribe-1-proを実行するように変更 +* [**Claude Sonnet 4.5の削除**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Anthropic Claudeノードから廃止 + +**新しいノード** +* [**Video Deflicker**](https://github.com/Comfy-Org/ComfyUI/pull/17754): ビデオのフリッカーを除去。より高速なRAFTオプティカルフローに対応 +* [**Split Image Channels**](https://github.com/Comfy-Org/ComfyUI/pull/13548): 1枚の画像から赤、緑、青、アルファの出力を生成 + + + + + +**パートナーノードの更新** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): 最大10枚の参照画像と長尺レイアウトに対応した画像生成と編集 +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): 1080pおよび2KでのReference to VideoとIdentity to Video + + + + + +**パートナーノードの更新** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): Bria 画像背景除去ノードに新しいモデルオプションを追加 + + + **パートナーノードの更新** diff --git a/.github/scripts/cms/staging/ko/changelog/index.mdx b/.github/scripts/cms/staging/ko/changelog/index.mdx index fec1827322..8dabf53824 100644 --- a/.github/scripts/cms/staging/ko/changelog/index.mdx +++ b/.github/scripts/cms/staging/ko/changelog/index.mdx @@ -4,6 +4,40 @@ cmsStaging: true --- + + +**신규 오픈소스 모델 지원** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): 코어에서 비디오 생성 모델 지원 +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): 사고, 인터리브 생성 및 자기 교정 기능과 함께 ModelSamplingSenseNova 노드 추가 +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): 텍스트 생성 노드에서 지원 + +**파트너 노드 업데이트** +* [**Kling 4.0 화면 비율**](https://github.com/Comfy-Org/ComfyUI/pull/17815): Kling 4.0 노드에 4:3 및 3:4 추가 +* [**Comfy Router의 Beeble 및 Higgsfield**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchX 및 Genjutsu 모션 전이를 Router 백엔드로 이동 +* [**Fish Audio 음성-텍스트 변환**](https://github.com/Comfy-Org/ComfyUI/pull/17828): 이제 transcribe-1-pro 실행 +* [**Claude Sonnet 4.5 제거**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Anthropic Claude 노드에서 지원 종료 + +**신규 노드** +* [**Video Deflicker**](https://github.com/Comfy-Org/ComfyUI/pull/17754): 비디오의 플리커를 제거하며, 더 빠른 RAFT 광학 흐름 사용 +* [**이미지 채널 분리**](https://github.com/Comfy-Org/ComfyUI/pull/13548): 하나의 이미지에서 빨간색, 초록색, 파란색 및 알파 출력 + + + + + +**파트너 노드 업데이트** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): 최대 10개의 레퍼런스 이미지를 활용한 이미지 생성 및 편집, 롱 레이아웃 +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): 1080p 및 2K 해상도의 레퍼런스 기반 비디오 생성 및 아이덴티티 기반 비디오 생성 + + + + + +**파트너 노드 업데이트** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): Bria 이미지 배경 제거 노드에 새로운 모델 옵션 추가 + + + **파트너 노드 업데이트** diff --git a/.github/scripts/cms/staging/ru/changelog/index.mdx b/.github/scripts/cms/staging/ru/changelog/index.mdx index 7777e52503..b822adf4d7 100644 --- a/.github/scripts/cms/staging/ru/changelog/index.mdx +++ b/.github/scripts/cms/staging/ru/changelog/index.mdx @@ -4,6 +4,40 @@ cmsStaging: true --- + + +**Поддержка новых open-source моделей** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): Поддержка модели генерации видео в ядре +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): Рассуждение, чередующаяся генерация и самокоррекция, добавлен узел ModelSamplingSenseNova +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): Поддерживается в узле Generate Text + +**Обновления партнёрских узлов** +* [**Соотношения сторон Kling 4.0**](https://github.com/Comfy-Org/ComfyUI/pull/17815): В узлы Kling 4.0 добавлены 4:3 и 3:4 +* [**Beeble и Higgsfield в Comfy Router**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchX и перенос движения Genjutsu переведены на бэкенд Router +* [**Речь в текст Fish Audio**](https://github.com/Comfy-Org/ComfyUI/pull/17828): Теперь работает на transcribe-1-pro +* [**Claude Sonnet 4.5 удалён**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Выведен из узла Anthropic Claude + +**Новые узлы** +* [**Video Deflicker**](https://github.com/Comfy-Org/ComfyUI/pull/17754): Устраняет мерцание видео с помощью более быстрого оптического потока RAFT +* [**Split Image Channels**](https://github.com/Comfy-Org/ComfyUI/pull/13548): Выходы красного, зелёного, синего и альфа-каналов из одного изображения + + + + + +**Обновления партнёрских узлов** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): Генерация и редактирование изображений с использованием до 10 референсных изображений, длинные макеты +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): Reference to Video и Identity to Video в разрешении 1080p и 2K + + + + + +**Обновления партнёрских нод** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): Новая опция модели в ноде Bria Remove Image Background + + + **Обновления партнёрских нод** diff --git a/.github/scripts/cms/staging/zh/changelog/index.mdx b/.github/scripts/cms/staging/zh/changelog/index.mdx index 81b071075f..e957106793 100644 --- a/.github/scripts/cms/staging/zh/changelog/index.mdx +++ b/.github/scripts/cms/staging/zh/changelog/index.mdx @@ -4,6 +4,40 @@ cmsStaging: true --- + + +**新增开源模型支持** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755):核心支持视频生成模型 +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032):思维、交错生成与自我纠正,新增 ModelSamplingSenseNova 节点 +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787):在 TextGenerate 节点中支持 + +**合作伙伴节点更新** +* [**Kling 4.0 宽高比**](https://github.com/Comfy-Org/ComfyUI/pull/17815):Kling 4.0 节点新增 4:3 和 3:4 +* [**Comfy Router 上的 Beeble 和 Higgsfield**](https://github.com/Comfy-Org/ComfyUI/pull/17794):SwitchX 与 Genjutsu 运动迁移已移至 Router 后端 +* [**Fish Audio 语音转文字**](https://github.com/Comfy-Org/ComfyUI/pull/17828):现在运行 transcribe-1-pro +* [**Claude Sonnet 4.5 已移除**](https://github.com/Comfy-Org/ComfyUI/pull/17819):已从 Anthropic Claude 节点停用 + +**新增节点** +* [**视频去闪烁**](https://github.com/Comfy-Org/ComfyUI/pull/17754):消除视频闪烁,采用更快的 RAFT 光流 +* [**拆分图像通道**](https://github.com/Comfy-Org/ComfyUI/pull/13548):从单张图像输出红、绿、蓝和透明度 + + + + + +**合作伙伴节点更新** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797):支持最多 10 张参考图像的图像生成与编辑,以及长版式布局 +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793):支持 1080p 和 2K 的参考图生视频与身份生视频 + + + + + +**合作伙伴节点更新** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780):Bria 移除图像背景节点新增模型选项 + + + **合作伙伴节点更新** diff --git a/changelog/index.mdx b/changelog/index.mdx index 53d11b52e0..61cd8cbfb4 100644 --- a/changelog/index.mdx +++ b/changelog/index.mdx @@ -4,6 +4,55 @@ description: "ComfyUI changelog: the latest features, improvements, and bug fixe icon: "clock-rotate-left" --- + + +**Partner Node Updates** +* [**Kling 4.0 aspect ratios**](https://github.com/Comfy-Org/ComfyUI/pull/17815): 4:3 and 3:4 added to the Kling 4.0 nodes +* [**Beeble and Higgsfield on Comfy Router**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchX and Genjutsu motion transfer moved to the Router backend +* [**Claude Sonnet 4.5 removed**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Retired from the Anthropic Claude node +* [**Fish Audio speech-to-text**](https://github.com/Comfy-Org/ComfyUI/pull/17828): Now runs transcribe-1-pro + +**New Open-Source Model Support** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): Video generation model support in core + +**New Nodes** +* [**Video Deflicker**](https://github.com/Comfy-Org/ComfyUI/pull/17754): Removes flicker from video, with a faster RAFT optical flow +* [**Split Image Channels**](https://github.com/Comfy-Org/ComfyUI/pull/13548): Red, green, blue and alpha outputs from one image + +**Improvements** +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): Thinking, interleaved generation and self-correction, plus the new ModelSamplingSenseNova node +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): Supported in Generate Text +* [**Save Image Advanced**](https://github.com/Comfy-Org/ComfyUI/pull/17727): Saves single channel PNGs +* [**Create Camera Info**](https://github.com/Comfy-Org/ComfyUI/pull/17730): Mesh input to frame any mesh +* [**Concatenate Video**](https://github.com/Comfy-Org/ComfyUI/pull/17621): Clips conform to the first clip's frame rate and audio +* [**--force-output-rerun**](https://github.com/Comfy-Org/ComfyUI/pull/17809): Output nodes write their files again on a cache hit +* [**Vocabulary-pruned Qwen checkpoints**](https://github.com/Comfy-Org/ComfyUI/pull/17633): Embedded tokenizers supported + +**Bug Fixes** +* [**Dynamic VRAM**](https://github.com/Comfy-Org/ComfyUI/pull/16491): Fixed a sampling schedule leaking between models +* [**APG**](https://github.com/Comfy-Org/ComfyUI/pull/17741): Fixed output changes at cfg 1.0 +* [**Google Omni**](https://github.com/Comfy-Org/ComfyUI/pull/17785): Video is downloaded when Google marks its file as failed + +**Removals** +* [**DirectML support removed**](https://github.com/Comfy-Org/ComfyUI/pull/17711): The --directml flag is gone, use CUDA or CPU + + + + + +**Partner Node Updates** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): Image generation and editing with up to 10 reference images, new long webpage and e-commerce layouts +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): Reference to Video and Identity to Video at 1080p and 2K + + + + + +**Partner Node Updates** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): RMBG 3.0 model option on the Bria Remove Image Background node + + + **Partner Node Updates** diff --git a/es/changelog/index.mdx b/es/changelog/index.mdx index f37a2b5649..8eddcaf034 100644 --- a/es/changelog/index.mdx +++ b/es/changelog/index.mdx @@ -2,9 +2,12 @@ title: "Registro de cambios" description: "Historial de ComfyUI: novedades, mejoras y correcciones de cada versión, actualizado a medida que se publican." icon: "clock-rotate-left" -translationSourceHash: 33458a4a +translationSourceHash: bdebb8af translationFrom: changelog/index.mdx translationBlockHashes: + "v0.46.0": 356cd787 + "v0.45.6": 03e2b995 + "v0.45.5": fc22765e "v0.45.4": 537893d0 "v0.45.3": dc0b0cd4 "v0.45.2": 932a7ad0 @@ -157,6 +160,55 @@ translationBlockHashes: "v0.3.40": 24608eaf --- + + +**Actualizaciones de nodos de socios** +* [**Relaciones de aspecto de Kling 4.0**](https://github.com/Comfy-Org/ComfyUI/pull/17815): se añaden 4:3 y 3:4 a los nodos de Kling 4.0 +* [**Beeble y Higgsfield en Comfy Router**](https://github.com/Comfy-Org/ComfyUI/pull/17794): la transferencia de movimiento de SwitchX y Genjutsu pasa al backend Router +* [**Claude Sonnet 4.5 eliminado**](https://github.com/Comfy-Org/ComfyUI/pull/17819): retirado del nodo Anthropic Claude +* [**Voz a texto de Fish Audio**](https://github.com/Comfy-Org/ComfyUI/pull/17828): ahora ejecuta transcribe-1-pro + +**Nuevo soporte de modelos de código abierto** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): soporte del modelo de generación de video en el núcleo + +**Nuevos nodos** +* [**Video Deflicker**](https://github.com/Comfy-Org/ComfyUI/pull/17754): elimina el parpadeo del video, con un flujo óptico RAFT más rápido +* [**Split Image Channels**](https://github.com/Comfy-Org/ComfyUI/pull/13548): salidas de rojo, verde, azul y alfa desde una sola imagen + +**Mejoras** +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): razonamiento, generación intercalada y autocorrección, además del nuevo nodo ModelSamplingSenseNova +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): compatible con Generar texto +* [**Guardar Imagen Avanzado**](https://github.com/Comfy-Org/ComfyUI/pull/17727): guarda PNG de un solo canal +* [**Crear información de cámara**](https://github.com/Comfy-Org/ComfyUI/pull/17730): entrada de malla para encuadrar cualquier malla +* [**Concatenar video**](https://github.com/Comfy-Org/ComfyUI/pull/17621): los clips se ajustan a la frecuencia de cuadros y al audio del primer clip +* [**--force-output-rerun**](https://github.com/Comfy-Org/ComfyUI/pull/17809): los nodos de salida vuelven a escribir sus archivos cuando hay un acierto de caché +* [**Checkpoints Qwen con vocabulario podado**](https://github.com/Comfy-Org/ComfyUI/pull/17633): se admiten tokenizers integrados + +**Correcciones de errores** +* [**VRAM dinámica**](https://github.com/Comfy-Org/ComfyUI/pull/16491): se corrigió una fuga de la programación de muestreo entre modelos +* [**APG**](https://github.com/Comfy-Org/ComfyUI/pull/17741): se corrigieron cambios en la salida a cfg 1.0 +* [**Google Omni**](https://github.com/Comfy-Org/ComfyUI/pull/17785): el video se descarga cuando Google marca su archivo como fallido + +**Eliminaciones** +* [**Se eliminó el soporte de DirectML**](https://github.com/Comfy-Org/ComfyUI/pull/17711): la marca --directml ya no existe, usa CUDA o CPU + + + + + +**Actualizaciones de nodos de socios** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): Generación y edición de imágenes con hasta 10 imágenes de referencia, nuevos diseños largos para páginas web y comercio electrónico +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): Reference to Video e Identity to Video en 1080p y 2K + + + + + +**Actualizaciones de nodos de socios** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): opción de modelo RMBG 3.0 en el nodo Bria: Quitar fondo de imagen + + + **Actualizaciones de nodos de socios** diff --git a/ja/changelog/index.mdx b/ja/changelog/index.mdx index 70be4269bf..846f7e0aaf 100644 --- a/ja/changelog/index.mdx +++ b/ja/changelog/index.mdx @@ -2,9 +2,12 @@ title: "変更履歴" description: "ComfyUI の変更ログ:各リリースの最新機能、改善点、バグ修正を、新バージョンのリリースに合わせて更新。" icon: "clock-rotate-left" -translationSourceHash: 33458a4a +translationSourceHash: bdebb8af translationFrom: changelog/index.mdx translationBlockHashes: + "v0.46.0": 356cd787 + "v0.45.6": 03e2b995 + "v0.45.5": fc22765e "v0.45.4": 537893d0 "v0.45.3": dc0b0cd4 "v0.45.2": 932a7ad0 @@ -157,6 +160,55 @@ translationBlockHashes: "v0.3.40": 24608eaf --- + + +**パートナーノードの更新** +* [**Kling 4.0 のアスペクト比**](https://github.com/Comfy-Org/ComfyUI/pull/17815): Kling 4.0 ノードに 4:3 と 3:4 を追加 +* [**Comfy Router 上の Beeble と Higgsfield**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchX と Genjutsu のモーショントランスファーを Router バックエンドへ移行 +* [**Claude Sonnet 4.5 を削除**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Anthropic Claude ノードから廃止 +* [**Fish Audio の音声テキスト変換**](https://github.com/Comfy-Org/ComfyUI/pull/17828): transcribe-1-pro で実行されるように + +**新しいオープンソースモデルのサポート** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): ビデオ生成モデルをコアでサポート + +**新しいノード** +* [**ビデオのフリッカー除去**](https://github.com/Comfy-Org/ComfyUI/pull/17754): ビデオのフリッカーを除去。RAFT オプティカルフローが高速化 +* [**画像チャンネルの分割**](https://github.com/Comfy-Org/ComfyUI/pull/13548): 1 枚の画像から赤、緑、青、アルファの出力を取得 + +**改善** +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): 思考、インターリーブ生成、自己修正に対応。さらに新しい ModelSamplingSenseNova ノードを追加 +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): TextGenerate でサポート +* [**画像を保存(高度な機能)**](https://github.com/Comfy-Org/ComfyUI/pull/17727): シングルチャンネルの PNG を保存 +* [**カメラ情報の作成**](https://github.com/Comfy-Org/ComfyUI/pull/17730): メッシュを入力して任意のメッシュにフレームを合わせる +* [**ビデオの連結**](https://github.com/Comfy-Org/ComfyUI/pull/17621): クリップが最初のクリップのフレームレートとオーディオに合わせられる +* [**--force-output-rerun**](https://github.com/Comfy-Org/ComfyUI/pull/17809): キャッシュヒット時に出力ノードがファイルを再度書き込む +* [**語彙を削減した Qwen チェックポイント**](https://github.com/Comfy-Org/ComfyUI/pull/17633): 埋め込み tokenizer をサポート + +**バグ修正** +* [**Dynamic VRAM**](https://github.com/Comfy-Org/ComfyUI/pull/16491): モデル間でサンプリングスケジュールが漏れ出す問題を修正 +* [**APG**](https://github.com/Comfy-Org/ComfyUI/pull/17741): cfg 1.0 で出力が変化する問題を修正 +* [**Google Omni**](https://github.com/Comfy-Org/ComfyUI/pull/17785): Google がファイルを失敗とマークした場合にビデオをダウンロードするように + +**削除** +* [**DirectML サポートの削除**](https://github.com/Comfy-Org/ComfyUI/pull/17711): --directml フラグは廃止。CUDA または CPU を使用してください + + + + + +**パートナーノードの更新** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): 最大10枚の参照画像を使用した画像生成と編集、新しい長尺ウェブページおよびeコマースレイアウト +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): 1080pおよび2KでのReference to VideoとIdentity to Video + + + + + +**パートナーノードの更新** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): Bria 画像背景除去ノードで RMBG 3.0 モデルオプションを選択可能 + + + **パートナーノードの更新** diff --git a/ko/changelog/index.mdx b/ko/changelog/index.mdx index 8a6648b1a0..93d6886403 100644 --- a/ko/changelog/index.mdx +++ b/ko/changelog/index.mdx @@ -2,9 +2,12 @@ title: "변경 로그" description: "ComfyUI 변경 로그: 각 릴리스의 최신 기능, 개선 사항, 버그 수정을 새 버전 출시에 맞춰 업데이트합니다." icon: "clock-rotate-left" -translationSourceHash: 33458a4a +translationSourceHash: bdebb8af translationFrom: changelog/index.mdx translationBlockHashes: + "v0.46.0": 356cd787 + "v0.45.6": 03e2b995 + "v0.45.5": fc22765e "v0.45.4": 537893d0 "v0.45.3": dc0b0cd4 "v0.45.2": 932a7ad0 @@ -157,6 +160,55 @@ translationBlockHashes: "v0.3.40": 24608eaf --- + + +**파트너 노드 업데이트** +* [**Kling 4.0 화면 비율**](https://github.com/Comfy-Org/ComfyUI/pull/17815): Kling 4.0 노드에 4:3과 3:4 추가 +* [**Comfy Router의 Beeble 및 Higgsfield**](https://github.com/Comfy-Org/ComfyUI/pull/17794): SwitchX와 Genjutsu 모션 전송이 Router 백엔드로 이동 +* [**Claude Sonnet 4.5 제거**](https://github.com/Comfy-Org/ComfyUI/pull/17819): Anthropic Claude 노드에서 제외됨 +* [**Fish Audio 음성-텍스트 변환**](https://github.com/Comfy-Org/ComfyUI/pull/17828): 이제 transcribe-1-pro를 실행 + +**새로운 오픈소스 모델 지원** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755): 코어에서 비디오 생성 모델 지원 + +**새 노드** +* [**비디오 플리커 제거**](https://github.com/Comfy-Org/ComfyUI/pull/17754): 비디오의 플리커를 제거하며, 더 빠른 RAFT 옵티컬 플로우를 사용 +* [**이미지 채널 분할**](https://github.com/Comfy-Org/ComfyUI/pull/13548): 하나의 이미지에서 빨간색, 초록색, 파란색, 알파 출력을 제공 + +**개선 사항** +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032): 사고, 인터리브 생성 및 자기 교정, 그리고 새로운 ModelSamplingSenseNova 노드 +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787): 텍스트 생성에서 지원 +* [**이미지 저장 고급**](https://github.com/Comfy-Org/ComfyUI/pull/17727): 단일 채널 PNG를 저장 +* [**카메라 정보 생성**](https://github.com/Comfy-Org/ComfyUI/pull/17730): 모든 메시의 프레임을 잡을 수 있는 메시 입력 +* [**비디오 연결**](https://github.com/Comfy-Org/ComfyUI/pull/17621): 클립이 첫 번째 클립의 프레임 레이트와 오디오에 맞춰짐 +* [**--force-output-rerun**](https://github.com/Comfy-Org/ComfyUI/pull/17809): 캐시 적중 시 출력 노드가 파일을 다시 기록 +* [**Vocabulary-pruned Qwen 체크포인트**](https://github.com/Comfy-Org/ComfyUI/pull/17633): 내장 tokenizer 지원 + +**버그 수정** +* [**동적 VRAM**](https://github.com/Comfy-Org/ComfyUI/pull/16491): 모델 간에 샘플링 스케줄이 누출되는 문제를 수정 +* [**APG**](https://github.com/Comfy-Org/ComfyUI/pull/17741): cfg 1.0에서 출력이 변경되는 문제를 수정 +* [**Google Omni**](https://github.com/Comfy-Org/ComfyUI/pull/17785): Google이 파일을 실패로 표시할 때 비디오를 다운로드 + +**제거** +* [**DirectML 지원 제거**](https://github.com/Comfy-Org/ComfyUI/pull/17711): --directml 플래그가 없어졌으므로 CUDA 또는 CPU를 사용하세요 + + + + + +**파트너 노드 업데이트** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797): 최대 10개의 레퍼런스 이미지를 사용한 이미지 생성 및 편집, 새로운 긴 웹페이지 및 이커머스 레이아웃 +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793): 1080p 및 2K 해상도의 레퍼런스 기반 비디오 생성과 Identity to Video + + + + + +**파트너 노드 업데이트** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780): Bria 이미지 배경 제거 노드에 RMBG 3.0 모델 옵션 추가 + + + **파트너 노드 업데이트** diff --git a/zh/changelog/index.mdx b/zh/changelog/index.mdx index d0f70bce95..d82cda25ea 100644 --- a/zh/changelog/index.mdx +++ b/zh/changelog/index.mdx @@ -2,9 +2,12 @@ title: "更新日志" description: "ComfyUI 变更日志:每个版本的最新功能、改进和错误修复,随新版本发布持续更新。" icon: "clock-rotate-left" -translationSourceHash: 33458a4a +translationSourceHash: bdebb8af translationFrom: changelog/index.mdx translationBlockHashes: + "v0.46.0": 356cd787 + "v0.45.6": 03e2b995 + "v0.45.5": fc22765e "v0.45.4": 537893d0 "v0.45.3": dc0b0cd4 "v0.45.2": 932a7ad0 @@ -157,6 +160,55 @@ translationBlockHashes: "v0.3.40": 24608eaf --- + + +**合作伙伴节点更新** +* [**Kling 4.0 宽高比**](https://github.com/Comfy-Org/ComfyUI/pull/17815):Kling 4.0 节点新增 4:3 和 3:4 +* [**Comfy Router 上的 Beeble 和 Higgsfield**](https://github.com/Comfy-Org/ComfyUI/pull/17794):SwitchX 和 Genjutsu 动作迁移已迁移至 Router 后端 +* [**Claude Sonnet 4.5 已移除**](https://github.com/Comfy-Org/ComfyUI/pull/17819):已从 Anthropic Claude 节点中停用 +* [**Fish Audio 语音转文字**](https://github.com/Comfy-Org/ComfyUI/pull/17828):现在运行 transcribe-1-pro + +**新增开源模型支持** +* [**PAN 2**](https://github.com/Comfy-Org/ComfyUI/pull/17755):核心中支持该视频生成模型 + +**新增节点** +* [**视频去闪烁**](https://github.com/Comfy-Org/ComfyUI/pull/17754):去除视频闪烁,并采用更快的 RAFT 光流 +* [**分割图像通道**](https://github.com/Comfy-Org/ComfyUI/pull/13548):从一张图像输出红色、绿色、蓝色和透明度通道 + +**改进** +* [**SenseNova**](https://github.com/Comfy-Org/ComfyUI/pull/16032):思考、交错生成与自我纠正,以及全新的 ModelSamplingSenseNova 节点 +* [**Qwen3-1.7B**](https://github.com/Comfy-Org/ComfyUI/pull/17787):TextGenerate 中已支持 +* [**保存图像高级**](https://github.com/Comfy-Org/ComfyUI/pull/17727):可保存单通道 PNG +* [**创建相机信息**](https://github.com/Comfy-Org/ComfyUI/pull/17730):通过网格输入为任意网格取景 +* [**拼接视频**](https://github.com/Comfy-Org/ComfyUI/pull/17621):片段会与第一个片段的帧率和音频保持一致 +* [**--force-output-rerun**](https://github.com/Comfy-Org/ComfyUI/pull/17809):命中缓存时输出节点会重新写入文件 +* [**词表裁剪的 Qwen 模型**](https://github.com/Comfy-Org/ComfyUI/pull/17633):支持内嵌 tokenizer + +**问题修复** +* [**动态 VRAM**](https://github.com/Comfy-Org/ComfyUI/pull/16491):修复了采样调度在模型之间泄漏的问题 +* [**APG**](https://github.com/Comfy-Org/ComfyUI/pull/17741):修复了 cfg 为 1.0 时输出发生变化的问题 +* [**Google Omni**](https://github.com/Comfy-Org/ComfyUI/pull/17785):当 Google 将文件标记为失败时会下载视频 + +**移除** +* [**移除 DirectML 支持**](https://github.com/Comfy-Org/ComfyUI/pull/17711):--directml 参数已移除,请使用 CUDA 或 CPU + + + + + +**合作伙伴节点更新** +* [**Kling 4.0 Image and Long Image**](https://github.com/Comfy-Org/ComfyUI/pull/17797):支持最多 10 张参考图像的图像生成与编辑,新增长网页和电商布局 +* [**HeyGen Avatar 6**](https://github.com/Comfy-Org/ComfyUI/pull/17793):参考视频生成(Reference to Video)与身份视频生成(Identity to Video),支持 1080p 和 2K + + + + + +**合作伙伴节点更新** +* [**Bria RMBG 3.0**](https://github.com/Comfy-Org/ComfyUI/pull/17780):Bria 移除图像背景节点上的 RMBG 3.0 模型选项 + + + **合作伙伴节点更新**