From 1fb49c1ee7aed4add80402d1a271dc9a7081686b Mon Sep 17 00:00:00 2001 From: lin-bot23 Date: Sat, 7 Nov 2026 13:05:00 +0800 Subject: [PATCH] docs(seedance): add 4K output for Seedance 2.5 and the 4K draft final --- .../bytedance/seedance-2-5-draft.mdx | 16 ++++++++-------- .../partner-nodes/bytedance/seedance-2-5.mdx | 10 +++++----- es/tutorials/partner-nodes/pricing.mdx | 8 ++++++-- .../bytedance/seedance-2-5-draft.mdx | 16 ++++++++-------- .../partner-nodes/bytedance/seedance-2-5.mdx | 10 +++++----- ja/tutorials/partner-nodes/pricing.mdx | 8 ++++++-- .../bytedance/seedance-2-5-draft.mdx | 16 ++++++++-------- .../partner-nodes/bytedance/seedance-2-5.mdx | 10 +++++----- ko/tutorials/partner-nodes/pricing.mdx | 8 ++++++-- .../bytedance/seedance-2-5-draft.mdx | 10 +++++----- .../partner-nodes/bytedance/seedance-2-5.mdx | 6 +++--- tutorials/partner-nodes/pricing.mdx | 4 ++++ .../bytedance/seedance-2-5-draft.mdx | 16 ++++++++-------- .../partner-nodes/bytedance/seedance-2-5.mdx | 10 +++++----- zh/tutorials/partner-nodes/pricing.mdx | 8 ++++++-- 15 files changed, 88 insertions(+), 68 deletions(-) diff --git a/es/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx b/es/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx index 92bcc15d8a..783c241ec4 100644 --- a/es/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx +++ b/es/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx @@ -1,12 +1,12 @@ --- -title: "Modo Draft de Seedance 2.5: vista previa a 480p y render final a 1080p" -description: "Itera con bajo costo en ComfyUI con el modo Draft de Seedance 2.5: vista previa a 480p con draft_task_id y render final a 1080p desde ese ID." +title: "Seedance 2.5 Draft: vista previa a 480p y final a 1080p o 4K" +description: "Itera con bajo costo en ComfyUI con el modo Draft de Seedance 2.5: vista previa a 480p y render final a 1080p o 4K desde su draft_task_id." sidebarTitle: "Seedance 2.5 Draft" -translationSourceHash: ca75e0fb +translationSourceHash: e8fc31cd translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx translationBlockHashes: - "_intro": 841ad114 - "How the two stages work": b0f496a5 + "_intro": 4a602410 + "How the two stages work": ca6bac84 "Workflow templates": e6b5f383 "Input materials": 370294e3 "Tips": 6f3c3468 @@ -15,7 +15,7 @@ import ReqHint from "/snippets/es/tutorials/partner-nodes/req-hint.mdx"; import UpdateReminder from "/snippets/es/tutorials/update-reminder.mdx"; -El modo Draft de Seedance 2.5 genera una vista previa rápida y de bajo costo en 480p antes de que te decidas por un render completo en 1080p, así puedes iterar sobre un prompt, un primer fotograma o un conjunto de imágenes de referencia sin pagar por generaciones a resolución completa. +El modo Draft de Seedance 2.5 genera una vista previa rápida y de bajo costo en 480p antes de que te decidas por un render completo en 1080p o 4K, así puedes iterar sobre un prompt, un primer fotograma o un conjunto de imágenes de referencia sin pagar por generaciones a resolución completa. El modo Draft está disponible en tres flujos de trabajo de Seedance 2.5. Selecciona **Seedance 2.5 Draft** como modelo en el nodo Text to Video, First-Last-Frame to Video o Reference to Video: @@ -23,7 +23,7 @@ El modo Draft está disponible en tres flujos de trabajo de Seedance 2.5. Selecc - **Imagen a video (I2V)**: itera sobre el primer fotograma - **Referencia a video (R2V)**: itera sobre las imágenes de referencia y el storyboard -La resolución está fijada en 480p en el modo Draft, y cada ejecución devuelve tanto la vista previa como un `draft_task_id` que usa el render final. +La resolución está fijada en 480p en el modo Draft, y cada ejecución devuelve tanto la vista previa como un `draft_task_id` que usa el render final. El nodo de render final ofrece 1080p y 4K, y 4K solo está disponible en cuentas empresariales. @@ -35,7 +35,7 @@ La resolución está fijada en 480p en el modo Draft, y cada ejecución devuelve Cada flujo de trabajo de borrador se ejecuta en dos etapas: 1. **Crear el borrador**: ejecuta el nodo Seedance 2.5 con el modelo **Seedance 2.5 Draft** para obtener la vista previa en 480p y su `draft_task_id`, que el nodo **Save Text** escribe en un archivo `.txt` en la carpeta de salida. Vuelve a ejecutar con un prompt o una entrada distintos hasta que el borrador se vea bien. -2. **Renderizar el video final**: copia el `draft_task_id` de ese archivo `.txt` en el campo `draft_task_id` del nodo **ByteDance Seedance 2.5 Draft to Final Video** y ejecútalo. El nodo renderiza el final en 1080p y reutiliza aquello con lo que se creó el borrador, incluidos el prompt, la entrada, la duración, la relación de aspecto y la configuración de audio, así que no hay que volver a introducir nada. +2. **Renderizar el video final**: copia el `draft_task_id` de ese archivo `.txt` en el campo `draft_task_id` del nodo **ByteDance Seedance 2.5 Draft to Final Video**, ajusta `resolution` a 1080p o 4K y ejecútalo. El nodo renderiza el video final y reutiliza aquello con lo que se creó el borrador, incluidos el prompt, la entrada, la duración, la relación de aspecto y la configuración de audio, así que no hay que volver a introducir nada. Cada plantilla coloca las dos etapas en grupos de nodos separados, llamados **Stage 1: Create and preview the draft video** y **Stage 2: Render the final video from the draft task ID**. Deja la etapa 1 en bypass cuando estés conforme con el borrador, porque una segunda ejecución del nodo de borrador crea un borrador nuevo en lugar de reutilizar el que revisaste. diff --git a/es/tutorials/partner-nodes/bytedance/seedance-2-5.mdx b/es/tutorials/partner-nodes/bytedance/seedance-2-5.mdx index f81caa4b63..b743e87f2c 100644 --- a/es/tutorials/partner-nodes/bytedance/seedance-2-5.mdx +++ b/es/tutorials/partner-nodes/bytedance/seedance-2-5.mdx @@ -2,13 +2,13 @@ title: "Seedance 2.5: generación de video con IA y audio" description: "Genera videos de 30 segundos con audio sincronizado, control de referencia multimodal y edición precisa con Seedance 2.5 en ComfyUI" sidebarTitle: "Seedance 2 5" -translationSourceHash: 4dcc9b5c +translationSourceHash: 66330855 translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5.mdx translationBlockHashes: "_intro": b6589680 "What Seedance 2.5 is good at": 5a91b792 "Example outputs": 23304d99 - "Available workflows": c27b5286 + "Available workflows": cf21df8c "Get started": 30f9b70a --- import ReqHint from "/snippets/es/tutorials/partner-nodes/req-hint.mdx"; @@ -45,7 +45,7 @@ Generación de imagen a video, animando una sola imagen hasta convertirla en un ### Texto a video (T2V) -Genera un video cinematográfico de 30 segundos a partir de un prompt de texto, con sonido y música sincronizados. El nodo admite resolución de salida de 480p, 720p y 1080p. +Genera un video cinematográfico de 30 segundos a partir de un prompt de texto, con sonido y música sincronizados. El nodo admite resolución de salida de 480p, 720p y 1080p, y 4K en cuentas empresariales. @@ -182,9 +182,9 @@ Descarga este video de entrada de ejemplo para probar el flujo de trabajo: -### Modo borrador (vista previa en 480p y luego final en 1080p) +### Modo borrador (vista previa en 480p y luego final en 1080p o 4K) -El modo borrador primero genera una vista previa en 480p de bajo costo y luego el resultado final en 1080p a partir de ese borrador. Se aplica a los flujos de trabajo Text to Video, First-Last-Frame to Video y Reference to Video. Consulta [Modo borrador de Seedance 2.5](/es/tutorials/partner-nodes/bytedance/seedance-2-5-draft) para ver las plantillas, las entradas de ejemplo y los consejos para el flujo de trabajo. +El modo borrador primero genera una vista previa en 480p de bajo costo y luego el resultado final en 1080p o 4K a partir de ese borrador. Se aplica a los flujos de trabajo Text to Video, First-Last-Frame to Video y Reference to Video. Consulta [Modo borrador de Seedance 2.5](/es/tutorials/partner-nodes/bytedance/seedance-2-5-draft) para ver las plantillas, las entradas de ejemplo y los consejos para el flujo de trabajo. ## Empezar diff --git a/es/tutorials/partner-nodes/pricing.mdx b/es/tutorials/partner-nodes/pricing.mdx index ce7abaf5ff..c971bca2ee 100644 --- a/es/tutorials/partner-nodes/pricing.mdx +++ b/es/tutorials/partner-nodes/pricing.mdx @@ -3,7 +3,7 @@ title: "Precios de los Nodos asociados" description: "Precios en créditos de los Nodos asociados de cada proveedor (Anthropic, OpenAI, Kling, Luma), por modelo, tipo de token y tipo de generación." sidebarTitle: "Precios" mode: "wide" -translationSourceHash: a08658db +translationSourceHash: 1091c8c7 translationFrom: tutorials/partner-nodes/pricing.mdx translationBlockHashes: "_intro": 1ec20604 @@ -11,7 +11,7 @@ translationBlockHashes: "Beeble": d1b19f61 "BFL": 7ae11dc6 "Bria": 94080005 - "ByteDance": d2af4769 + "ByteDance": e1a9d40e "Clarity AI": 3d568b52 "ElevenLabs": cadb48cc "FishAudio": c957b7da @@ -262,9 +262,13 @@ Facturación basada en tokens. Consumo estimado de tokens = **(duración del vid | Seedance 2.5 | 480p, 720p | sí | 1.931 | | Seedance 2.5 | 1080p | no | 3.530 | | Seedance 2.5 | 1080p | sí | 2.112 | +| Seedance 2.5 | 4K | no | 1.418 | +| Seedance 2.5 | 4K | sí | 0.875 | **Entrada de video** significa que hay un video de referencia conectado. Las referencias de texto, imagen y audio cuentan como entrada sin video. Hay límites mínimos de consumo de tokens cuando la entrada incluye video: si el consumo estimado es inferior al límite, la facturación aplica el mínimo. +La salida 4K solo está disponible en cuentas empresariales. Un fotograma 4K contiene unos cuatro veces más tokens que uno de 1080p con la misma duración, así que el precio por 1K tokens es menor pero el cargo total es mayor. + Los videos de referencia deben durar al menos 2 segundos y la entrada total está limitada a 30 segundos. Cargos estimados para una salida en 16:9 a 24 fps: | Resolución | Salida de 5s, sin entrada de video | Salida de 5s + entrada de video de 2–30s | diff --git a/ja/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx b/ja/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx index b0da581521..358d11e80d 100644 --- a/ja/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx +++ b/ja/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx @@ -1,12 +1,12 @@ --- -title: "Seedance 2.5 Draftモード:480pでプレビューし、1080pでレンダリング" -description: "ComfyUIのSeedance 2.5 Draftモードで低コストに反復:draft_task_idで480pのプレビューをレンダリングし、そのIDから1080pの最終版をレンダリングします。" +title: "Seedance 2.5 Draftモード:480pでプレビューし、1080pまたは4Kでレンダリング" +description: "ComfyUIのSeedance 2.5 Draftモードで低コストに反復:480pのプレビューをレンダリングし、そのdraft_task_idから1080pまたは4Kの最終版をレンダリングします。" sidebarTitle: "Seedance 2.5 Draft" -translationSourceHash: ca75e0fb +translationSourceHash: e8fc31cd translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx translationBlockHashes: - "_intro": 841ad114 - "How the two stages work": b0f496a5 + "_intro": 4a602410 + "How the two stages work": ca6bac84 "Workflow templates": e6b5f383 "Input materials": 370294e3 "Tips": 6f3c3468 @@ -16,7 +16,7 @@ import ReqHint from "/snippets/ja/tutorials/partner-nodes/req-hint.mdx"; import UpdateReminder from "/snippets/ja/tutorials/update-reminder.mdx"; -Seedance 2.5 のドラフトモードでは、フル解像度の 1080p レンダリングを実行する前に、高速で低コストな 480p プレビューをレンダリングできます。そのため、フル解像度の生成コストをかけずに、プロンプト、先頭フレーム、または参照画像のセットを試行錯誤できます。 +Seedance 2.5 のドラフトモードでは、フル解像度の 1080p または 4K レンダリングを実行する前に、高速で低コストな 480p プレビューをレンダリングできます。そのため、フル解像度の生成コストをかけずに、プロンプト、先頭フレーム、または参照画像のセットを試行錯誤できます。 ドラフトモードは 3 つの Seedance 2.5 ワークフローで利用できます。Text to Video、First-Last-Frame to Video、または Reference to Video ノードでモデルとして **Seedance 2.5 Draft** を選択します: @@ -24,7 +24,7 @@ Seedance 2.5 のドラフトモードでは、フル解像度の 1080p レンダ - **Image to video (I2V)**: 先頭フレームを試行錯誤する - **Reference to video (R2V)**: 参照画像とストーリーボードを試行錯誤する -ドラフトモードでは解像度は 480p に固定され、各実行ではプレビューと、最終レンダリングで使用される `draft_task_id` の両方が返されます。 +ドラフトモードでは解像度は 480p に固定され、各実行ではプレビューと、最終レンダリングで使用される `draft_task_id` の両方が返されます。最終レンダリング用ノードでは1080pと4Kを選択でき、4Kはエンタープライズアカウントのみ利用できます。 @@ -36,7 +36,7 @@ Seedance 2.5 のドラフトモードでは、フル解像度の 1080p レンダ すべてのドラフトワークフローは 2 つのステージで実行されます: 1. **ドラフトを作成**: **Seedance 2.5 Draft** モデルを使って Seedance 2.5 ノードを実行し、480p のプレビューとその `draft_task_id` を取得します。`draft_task_id` は **Save Text** ノードによって出力フォルダー内の `.txt` ファイルに書き込まれます。ドラフトが意図どおりになるまで、別のプロンプトや入力で再実行してください。 -2. **最終版をレンダリング**: その `.txt` ファイルから `draft_task_id` をコピーし、**ByteDance Seedance 2.5 Draft to Final Video** ノードの `draft_task_id` フィールドに入力して実行します。このノードは 1080p の最終版をレンダリングし、プロンプト、入力、再生時間、アスペクト比、オーディオ設定など、ドラフトの作成時に使われたものを再利用するため、再入力するものはありません。 +2. **最終版をレンダリング**: その `.txt` ファイルから `draft_task_id` をコピーし、**ByteDance Seedance 2.5 Draft to Final Video** ノードの `draft_task_id` フィールドに入力し、`resolution` を1080pまたは4Kに設定して実行します。このノードは最終版をレンダリングし、プロンプト、入力、再生時間、アスペクト比、オーディオ設定など、ドラフトの作成時に使われたものを再利用するため、再入力するものはありません。 各テンプレートは 2 つのステージを別々のノードグループに配置し、それぞれ **Stage 1: Create and preview the draft video** と **Stage 2: Render the final video from the draft task ID** という名前が付けられています。ドラフトに満足したら、Stage 1 はバイパスしたままにしてください。ドラフトノードをもう一度実行すると、確認したドラフトを再利用するのではなく、新しいドラフトが作成されるためです。 diff --git a/ja/tutorials/partner-nodes/bytedance/seedance-2-5.mdx b/ja/tutorials/partner-nodes/bytedance/seedance-2-5.mdx index d7e2614ef6..1476a63a2b 100644 --- a/ja/tutorials/partner-nodes/bytedance/seedance-2-5.mdx +++ b/ja/tutorials/partner-nodes/bytedance/seedance-2-5.mdx @@ -2,13 +2,13 @@ title: "Seedance 2.5 - オーディオ付きAIビデオ生成" description: "ComfyUIでSeedance 2.5を使用して、同期オーディオ、マルチモーダル参照制御、精密な編集に対応した30秒のビデオを生成します" sidebarTitle: "Seedance 2.5" -translationSourceHash: 4dcc9b5c +translationSourceHash: 66330855 translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5.mdx translationBlockHashes: "_intro": b6589680 "What Seedance 2.5 is good at": 5a91b792 "Example outputs": 23304d99 - "Available workflows": c27b5286 + "Available workflows": cf21df8c "Get started": 30f9b70a --- @@ -47,7 +47,7 @@ Seedance 2.5 は、2026年7月31日にリリースされた ByteDance の新世 ### テキストから動画生成(T2V) -テキストプロンプトから、同期されたサウンドと音楽を備えた30秒のシネマティックなビデオを生成します。ノードは480p、720p、1080pの出力解像度に対応しています。 +テキストプロンプトから、同期されたサウンドと音楽を備えた30秒のシネマティックなビデオを生成します。ノードは480p、720p、1080pの出力解像度に対応し、エンタープライズアカウントでは4K出力も利用できます。 @@ -184,9 +184,9 @@ Seedance 2.5 は、2026年7月31日にリリースされた ByteDance の新世 -### ドラフトモード(480pプレビューから1080p最終版へ) +### ドラフトモード(480pプレビューから1080pまたは4K最終版へ) -ドラフトモードは、まず低コストの 480p プレビューをレンダリングし、そのドラフトから 1080p の最終版をレンダリングします。テキストから動画生成、先頭・末尾フレームから動画生成、参照から動画生成の各ワークフローに適用されます。テンプレート、サンプル入力、ワークフローのヒントについては、[Seedance 2.5 ドラフトモード](/ja/tutorials/partner-nodes/bytedance/seedance-2-5-draft)を参照してください。 +ドラフトモードは、まず低コストの 480p プレビューをレンダリングし、そのドラフトから 1080p または 4K の最終版をレンダリングします。テキストから動画生成、先頭・末尾フレームから動画生成、参照から動画生成の各ワークフローに適用されます。テンプレート、サンプル入力、ワークフローのヒントについては、[Seedance 2.5 ドラフトモード](/ja/tutorials/partner-nodes/bytedance/seedance-2-5-draft)を参照してください。 ## はじめる diff --git a/ja/tutorials/partner-nodes/pricing.mdx b/ja/tutorials/partner-nodes/pricing.mdx index fb774fcbfb..beca4a3ef3 100644 --- a/ja/tutorials/partner-nodes/pricing.mdx +++ b/ja/tutorials/partner-nodes/pricing.mdx @@ -3,7 +3,7 @@ title: "Partner Node 価格" description: "Anthropic、OpenAI、Kling、Luma など全プロバイダーの Partner Node クレジット価格。モデル、トークン種別、生成種別ごとに分類。" sidebarTitle: "料金" mode: "wide" -translationSourceHash: a08658db +translationSourceHash: 1091c8c7 translationFrom: tutorials/partner-nodes/pricing.mdx translationBlockHashes: "_intro": 1ec20604 @@ -11,7 +11,7 @@ translationBlockHashes: "Beeble": d1b19f61 "BFL": 7ae11dc6 "Bria": 94080005 - "ByteDance": d2af4769 + "ByteDance": e1a9d40e "Clarity AI": 3d568b52 "ElevenLabs": cadb48cc "FishAudio": c957b7da @@ -262,9 +262,13 @@ Seedream 5.0 Proは解像度ベースの課金方式を採用しています:* | Seedance 2.5 | 480p, 720p | あり | 1.931 | | Seedance 2.5 | 1080p | なし | 3.530 | | Seedance 2.5 | 1080p | あり | 2.112 | +| Seedance 2.5 | 4K | なし | 1.418 | +| Seedance 2.5 | 4K | あり | 0.875 | **動画入力**は、参照動画が接続されていることを意味します。テキスト、画像、音声の参照は動画入力なしとしてカウントされます。入力に動画が含まれる場合、最低トークン消費量の制限が適用されます:推定消費量が最低値を下回る場合、最低消費量で課金されます。 +4K出力はエンタープライズアカウントのみ利用できます。同じ長さでは4Kの1フレームあたりのトークン数が1080pの約4倍になるため、1Kトークンあたりの単価は低くなりますが、合計料金は高くなります。 + 参照動画は2秒以上、入力の合計は最大30秒です。16:9出力(24fps)の推定料金: | 解像度 | 5秒出力、動画入力なし | 5秒出力 + 2〜30秒の動画入力 | diff --git a/ko/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx b/ko/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx index 9bbb5a8fbd..0948d6c232 100644 --- a/ko/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx +++ b/ko/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx @@ -1,12 +1,12 @@ --- -title: "Seedance 2.5 Draft 모드: 480p로 미리보기 후 1080p로 렌더링" -description: "ComfyUI의 Seedance 2.5 Draft 모드로 저렴하게 반복 작업하세요: draft_task_id로 480p 미리보기를 렌더링한 다음, 그 ID로 1080p 최종본을 렌더링합니다." +title: "Seedance 2.5 Draft 모드: 480p로 미리보기 후 1080p 또는 4K로 렌더링" +description: "ComfyUI의 Seedance 2.5 Draft 모드로 저렴하게 반복 작업하세요: 480p 미리보기를 렌더링한 다음, 그 draft_task_id로 1080p 또는 4K 최종본을 렌더링합니다." sidebarTitle: "Seedance 2.5 Draft" -translationSourceHash: ca75e0fb +translationSourceHash: e8fc31cd translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx translationBlockHashes: - "_intro": 841ad114 - "How the two stages work": b0f496a5 + "_intro": 4a602410 + "How the two stages work": ca6bac84 "Workflow templates": e6b5f383 "Input materials": 370294e3 "Tips": 6f3c3468 @@ -16,7 +16,7 @@ import ReqHint from "/snippets/ko/tutorials/partner-nodes/req-hint.mdx"; import UpdateReminder from "/snippets/ko/tutorials/update-reminder.mdx"; -Seedance 2.5 Draft 모드는 완전한 1080p 렌더를 확정하기 전에 빠르고 저렴한 480p 미리보기를 렌더링하므로, 전체 해상도 생성 비용을 치르지 않고도 프롬프트, 첫 프레임 또는 레퍼런스 이미지 세트를 반복 개선할 수 있습니다. +Seedance 2.5 Draft 모드는 완전한 1080p 또는 4K 렌더를 확정하기 전에 빠르고 저렴한 480p 미리보기를 렌더링하므로, 전체 해상도 생성 비용을 치르지 않고도 프롬프트, 첫 프레임 또는 레퍼런스 이미지 세트를 반복 개선할 수 있습니다. Draft 모드는 세 가지 Seedance 2.5 워크플로에서 사용할 수 있습니다. 텍스트 기반 비디오 생성(Text to Video), 첫/마지막 프레임 기반 비디오 생성(First-Last-Frame to Video), 레퍼런스 기반 비디오 생성(Reference to Video) 노드에서 모델로 **Seedance 2.5 Draft**를 선택하세요: @@ -24,7 +24,7 @@ Draft 모드는 세 가지 Seedance 2.5 워크플로에서 사용할 수 있습 - **이미지 기반 비디오 생성(I2V)**: 첫 프레임을 반복 조정 - **레퍼런스 기반 비디오 생성(R2V)**: 레퍼런스 이미지와 스토리보드를 반복 조정 -Draft 모드에서는 해상도가 480p로 고정되며, 각 실행은 미리보기와 함께 최종 렌더에서 사용하는 `draft_task_id`를 반환합니다. +Draft 모드에서는 해상도가 480p로 고정되며, 각 실행은 미리보기와 함께 최종 렌더에서 사용하는 `draft_task_id`를 반환합니다. 최종 렌더 노드에서는 1080p와 4K를 선택할 수 있으며 4K는 엔터프라이즈 계정에서만 사용할 수 있습니다. @@ -36,7 +36,7 @@ Draft 모드에서는 해상도가 480p로 고정되며, 각 실행은 미리보 모든 초안 워크플로는 두 단계로 실행됩니다: 1. **초안 생성**: **Seedance 2.5 Draft** 모델로 Seedance 2.5 노드를 실행하여 480p 미리보기와 해당 `draft_task_id`를 얻습니다. 이 ID는 **Save Text** 노드가 출력 폴더의 `.txt` 파일에 기록합니다. 초안이 원하는 대로 나올 때까지 다른 프롬프트나 입력으로 다시 실행하세요. -2. **최종 렌더링**: 해당 `.txt` 파일에서 `draft_task_id`를 복사하여 **ByteDance Seedance 2.5 Draft to Final Video** 노드의 `draft_task_id` 필드에 넣고 실행하세요. 이 노드는 1080p 최종 영상을 렌더링하면서 초안을 생성할 때 사용한 프롬프트, 입력, 재생 시간, 화면 비율, 오디오 설정을 그대로 재사용하므로 다시 입력할 내용이 없습니다. +2. **최종 렌더링**: 해당 `.txt` 파일에서 `draft_task_id`를 복사하여 **ByteDance Seedance 2.5 Draft to Final Video** 노드의 `draft_task_id` 필드에 넣고 `resolution`을 1080p 또는 4K로 설정한 뒤 실행하세요. 이 노드는 최종 영상을 렌더링하면서 초안을 생성할 때 사용한 프롬프트, 입력, 재생 시간, 화면 비율, 오디오 설정을 그대로 재사용하므로 다시 입력할 내용이 없습니다. 각 템플릿은 두 단계를 별도의 노드 그룹에 배치하며, 이름은 **Stage 1: Create and preview the draft video**(1단계: 초안 비디오 생성 및 미리보기)와 **Stage 2: Render the final video from the draft task ID**(2단계: draft task ID로 최종 비디오 렌더링)입니다. 초안이 마음에 들면 1단계는 비활성화(bypass)해 두세요. 초안 노드를 다시 실행하면 검토한 초안을 재사용하는 대신 새로운 초안이 생성되기 때문입니다. diff --git a/ko/tutorials/partner-nodes/bytedance/seedance-2-5.mdx b/ko/tutorials/partner-nodes/bytedance/seedance-2-5.mdx index 80e5d2ca15..585d22ab96 100644 --- a/ko/tutorials/partner-nodes/bytedance/seedance-2-5.mdx +++ b/ko/tutorials/partner-nodes/bytedance/seedance-2-5.mdx @@ -2,13 +2,13 @@ title: "Seedance 2.5: 오디오가 포함된 AI 비디오 생성" description: "ComfyUI에서 Seedance 2.5를 사용하여 동기화된 오디오, 멀티모달 참조 제어, 정밀한 편집 기능으로 30초 비디오를 생성하세요" sidebarTitle: "Seedance 2.5" -translationSourceHash: 4dcc9b5c +translationSourceHash: 66330855 translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5.mdx translationBlockHashes: "_intro": b6589680 "What Seedance 2.5 is good at": 5a91b792 "Example outputs": 23304d99 - "Available workflows": c27b5286 + "Available workflows": cf21df8c "Get started": 30f9b70a --- @@ -46,7 +46,7 @@ Seedance 2.5는 2026년 7월 31일에 출시된 ByteDance의 차세대 비디오 ### 텍스트 기반 비디오 생성 (T2V) -텍스트 프롬프트에서 동기화된 사운드와 음악이 포함된 30초 시네마틱 비디오를 생성합니다. 노드는 480p, 720p, 1080p 출력 해상도를 지원합니다. +텍스트 프롬프트에서 동기화된 사운드와 음악이 포함된 30초 시네마틱 비디오를 생성합니다. 노드는 480p, 720p, 1080p 출력 해상도를 지원하며, 엔터프라이즈 계정에서는 4K 출력도 사용할 수 있습니다. @@ -183,9 +183,9 @@ Seedance 2.5는 2026년 7월 31일에 출시된 ByteDance의 차세대 비디오 -### 드래프트 모드 (480p 미리보기 후 1080p 최종본) +### 드래프트 모드 (480p 미리보기 후 1080p 또는 4K 최종본) -드래프트 모드는 먼저 저비용 480p 미리보기를 렌더링한 다음, 그 드래프트를 기반으로 1080p 최종본을 렌더링합니다. 텍스트 기반 비디오 생성, 첫/마지막 프레임 기반 비디오 생성, 레퍼런스 기반 비디오 생성 워크플로에 적용됩니다. 템플릿, 샘플 입력, 워크플로 팁은 [Seedance 2.5 드래프트 모드](/ko/tutorials/partner-nodes/bytedance/seedance-2-5-draft)를 참조하세요. +드래프트 모드는 먼저 저비용 480p 미리보기를 렌더링한 다음, 그 드래프트를 기반으로 1080p 또는 4K 최종본을 렌더링합니다. 텍스트 기반 비디오 생성, 첫/마지막 프레임 기반 비디오 생성, 레퍼런스 기반 비디오 생성 워크플로에 적용됩니다. 템플릿, 샘플 입력, 워크플로 팁은 [Seedance 2.5 드래프트 모드](/ko/tutorials/partner-nodes/bytedance/seedance-2-5-draft)를 참조하세요. ## 시작하기 diff --git a/ko/tutorials/partner-nodes/pricing.mdx b/ko/tutorials/partner-nodes/pricing.mdx index 3c1c69bb4a..b5b293361f 100644 --- a/ko/tutorials/partner-nodes/pricing.mdx +++ b/ko/tutorials/partner-nodes/pricing.mdx @@ -3,7 +3,7 @@ title: "Partner Node 가격" description: "Anthropic, OpenAI, Kling, Luma 등 모든 제공업체의 Partner Node 크레딧 가격을 모델, 토큰 유형, 생성 유형별로 구분합니다." sidebarTitle: "가격 정책" mode: "wide" -translationSourceHash: a08658db +translationSourceHash: 1091c8c7 translationFrom: tutorials/partner-nodes/pricing.mdx translationBlockHashes: "_intro": 1ec20604 @@ -11,7 +11,7 @@ translationBlockHashes: "Beeble": d1b19f61 "BFL": 7ae11dc6 "Bria": 94080005 - "ByteDance": d2af4769 + "ByteDance": e1a9d40e "Clarity AI": 3d568b52 "ElevenLabs": cadb48cc "FishAudio": c957b7da @@ -262,9 +262,13 @@ Seedream 5.0 Pro는 해상도 기반 요금제를 사용합니다: **1K** (≤2. | Seedance 2.5 | 480p, 720p | 예 | 1.931 | | Seedance 2.5 | 1080p | 아니요 | 3.530 | | Seedance 2.5 | 1080p | 예 | 2.112 | +| Seedance 2.5 | 4K | 아니요 | 1.418 | +| Seedance 2.5 | 4K | 예 | 0.875 | **비디오 입력**은 참조 비디오가 연결됨을 의미합니다. 텍스트, 이미지 및 오디오 참조는 비디오 입력 없음으로 간주됩니다. 입력에 비디오가 포함된 경우 최소 토큰 소비량 제한이 적용됩니다: 예상 소비량이 최소값보다 낮으면 최소 소비량으로 청구됩니다. +4K 출력은 엔터프라이즈 계정에서만 사용할 수 있습니다. 같은 길이에서 4K 프레임은 1080p 프레임보다 토큰이 약 4배 많으므로 1K 토큰당 단가는 낮지만 총 비용은 더 높습니다. + 참조 비디오는 최소 2초 이상이어야 하며, 입력 합계는 최대 30초입니다. 16:9 출력(24fps)의 예상 요금: | 해상도 | 5초 출력, 비디오 입력 없음 | 5초 출력 + 2–30초 비디오 입력 | diff --git a/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx b/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx index 5475713fed..b46759bec5 100644 --- a/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx +++ b/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx @@ -1,13 +1,13 @@ --- -title: "Seedance 2.5 Draft mode: preview at 480p, then render at 1080p" -description: "Iterate cheaply with Seedance 2.5 Draft mode in ComfyUI: render a 480p preview with a draft_task_id, then render the 1080p final from that ID." +title: "Seedance 2.5 Draft mode: preview at 480p, then render at 1080p or 4K" +description: "Iterate cheaply with Seedance 2.5 Draft mode in ComfyUI: render a 480p preview, then render the 1080p or 4K final from its draft_task_id." sidebarTitle: "Seedance 2.5 Draft" --- import ReqHint from "/snippets/tutorials/partner-nodes/req-hint.mdx"; import UpdateReminder from "/snippets/tutorials/update-reminder.mdx"; -Seedance 2.5 Draft mode renders a fast, low-cost 480p preview before you commit to a full 1080p render, so you can iterate on a prompt, a first frame, or a set of reference images without paying for full-resolution generations. +Seedance 2.5 Draft mode renders a fast, low-cost 480p preview before you commit to a full 1080p or 4K render, so you can iterate on a prompt, a first frame, or a set of reference images without paying for full-resolution generations. Draft mode is available on three Seedance 2.5 workflows. Select **Seedance 2.5 Draft** as the model on the Text to Video, First-Last-Frame to Video, or Reference to Video node: @@ -15,7 +15,7 @@ Draft mode is available on three Seedance 2.5 workflows. Select **Seedance 2.5 D - **Image to video (I2V)**: iterate on the first frame - **Reference to video (R2V)**: iterate on the reference images and the storyboard -Resolution is locked to 480p in draft mode, and each run returns both the preview and a `draft_task_id` that the final render uses. +Resolution is locked to 480p in draft mode, and each run returns both the preview and a `draft_task_id` that the final render uses. The final render node offers 1080p and 4K, and 4K is available to enterprise accounts only. @@ -27,7 +27,7 @@ Resolution is locked to 480p in draft mode, and each run returns both the previe Every draft workflow runs in two stages: 1. **Create the draft**: run the Seedance 2.5 node with the **Seedance 2.5 Draft** model to get the 480p preview and its `draft_task_id`, which the **Save Text** node writes to a `.txt` file in the output folder. Rerun with a different prompt or input until the draft looks right. -2. **Render the final**: copy the `draft_task_id` from that `.txt` file into the `draft_task_id` field on the **ByteDance Seedance 2.5 Draft to Final Video** node and run it. The node renders the 1080p final and reuses what the draft was created with, including the prompt, the input, the duration, the aspect ratio, and the audio setting, so there is nothing to re-enter. +2. **Render the final**: copy the `draft_task_id` from that `.txt` file into the `draft_task_id` field on the **ByteDance Seedance 2.5 Draft to Final Video** node, set `resolution` to 1080p or 4K, and run it. The node renders the final and reuses what the draft was created with, including the prompt, the input, the duration, the aspect ratio, and the audio setting, so there is nothing to re-enter. Each template puts the two stages in separate node groups, named **Stage 1: Create and preview the draft video** and **Stage 2: Render the final video from the draft task ID**. Leave Stage 1 bypassed once you are happy with the draft, because a second run of the draft node creates a new draft instead of reusing the one you reviewed. diff --git a/tutorials/partner-nodes/bytedance/seedance-2-5.mdx b/tutorials/partner-nodes/bytedance/seedance-2-5.mdx index 2a4c564970..448a41e448 100644 --- a/tutorials/partner-nodes/bytedance/seedance-2-5.mdx +++ b/tutorials/partner-nodes/bytedance/seedance-2-5.mdx @@ -37,7 +37,7 @@ Image-to-video generation, animating a single image into a 1080p clip with sound ### Text to video (T2V) -Generate a 30-second cinematic video from a text prompt, with synchronized sound and music. The node supports 480p, 720p, and 1080p output resolution. +Generate a 30-second cinematic video from a text prompt, with synchronized sound and music. The node supports 480p, 720p, and 1080p output resolution, plus 4K output on enterprise accounts. @@ -174,9 +174,9 @@ Download this sample input video to try the workflow: -### Draft mode (480p preview, then 1080p final) +### Draft mode (480p preview, then 1080p or 4K final) -Draft mode renders a low-cost 480p preview first, then the 1080p final from that draft. It applies to the Text to Video, First-Last-Frame to Video, and Reference to Video workflows. See [Seedance 2.5 Draft mode](/tutorials/partner-nodes/bytedance/seedance-2-5-draft) for the templates, sample inputs, and workflow tips. +Draft mode renders a low-cost 480p preview first, then the 1080p or 4K final from that draft. It applies to the Text to Video, First-Last-Frame to Video, and Reference to Video workflows. See [Seedance 2.5 Draft mode](/tutorials/partner-nodes/bytedance/seedance-2-5-draft) for the templates, sample inputs, and workflow tips. ## Get started diff --git a/tutorials/partner-nodes/pricing.mdx b/tutorials/partner-nodes/pricing.mdx index 2144a2bc5a..e7e288e4a7 100644 --- a/tutorials/partner-nodes/pricing.mdx +++ b/tutorials/partner-nodes/pricing.mdx @@ -213,9 +213,13 @@ Token-based billing. Estimated token consumption = **(input video duration + out | Seedance 2.5 | 480p, 720p | yes | 1.931 | | Seedance 2.5 | 1080p | no | 3.530 | | Seedance 2.5 | 1080p | yes | 2.112 | +| Seedance 2.5 | 4K | no | 1.418 | +| Seedance 2.5 | 4K | yes | 0.875 | **Video input** means a reference video is connected. Text, image, and audio references count as no video input. Minimum token consumption limits apply when the input includes video: if the estimated consumption is below the limit, billing uses the minimum. +4K output is available to enterprise accounts only. A 4K frame carries about four times the tokens of a 1080p frame at the same duration, so the per-1K-token rate is lower while the total charge is higher. + Reference videos must be at least 2 seconds long, and total input is limited to 30 seconds. Estimated charges for 16:9 output at 24 fps: | Resolution | 5s output, no video input | 5s output + 2–30s video input | diff --git a/zh/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx b/zh/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx index cc28e275d5..5f0ea304c7 100644 --- a/zh/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx +++ b/zh/tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx @@ -1,12 +1,12 @@ --- -title: "Seedance 2.5 草稿模式:先以 480p 预览,再渲染 1080p" -description: "在 ComfyUI 中使用 Seedance 2.5 草稿模式低成本迭代:通过 draft_task_id 渲染 480p 预览,再根据该 ID 渲染 1080p 最终成品。" +title: "Seedance 2.5 草稿模式:先以 480p 预览,再渲染 1080p 或 4K" +description: "在 ComfyUI 中使用 Seedance 2.5 草稿模式低成本迭代:先渲染 480p 预览,再根据其 draft_task_id 渲染 1080p 或 4K 最终成品。" sidebarTitle: "Seedance 2.5 草稿" -translationSourceHash: ca75e0fb +translationSourceHash: e8fc31cd translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5-draft.mdx translationBlockHashes: - "_intro": 841ad114 - "How the two stages work": b0f496a5 + "_intro": 4a602410 + "How the two stages work": ca6bac84 "Workflow templates": e6b5f383 "Input materials": 370294e3 "Tips": 6f3c3468 @@ -16,7 +16,7 @@ import ReqHint from "/snippets/zh/tutorials/partner-nodes/req-hint.mdx"; import UpdateReminder from "/snippets/zh/tutorials/update-reminder.mdx"; -Seedance 2.5 Draft 模式会先渲染一个快速、低成本的 480p 预览,让你在确定投入完整 1080p 渲染之前,就能反复调整提示词、首帧或一组参考图像,而无需为全分辨率生成付费。 +Seedance 2.5 Draft 模式会先渲染一个快速、低成本的 480p 预览,让你在确定投入完整 1080p 或 4K 渲染之前,就能反复调整提示词、首帧或一组参考图像,而无需为全分辨率生成付费。 Draft 模式可在三个 Seedance 2.5 工作流中使用。在文本到视频、首尾帧到视频或参考到视频节点上,将模型选择为 **Seedance 2.5 Draft**: @@ -24,7 +24,7 @@ Draft 模式可在三个 Seedance 2.5 工作流中使用。在文本到视频、 - **图像到视频 (I2V)**:反复调整首帧 - **参考到视频 (R2V)**:反复调整参考图像和故事板 -在 Draft 模式下,分辨率固定为 480p,每次运行都会同时返回预览和一个 `draft_task_id`,供最终渲染使用。 +在 Draft 模式下,分辨率固定为 480p,每次运行都会同时返回预览和一个 `draft_task_id`,供最终渲染使用。最终渲染节点提供 1080p 和 4K,其中 4K 仅企业账号可用。 @@ -36,7 +36,7 @@ Draft 模式可在三个 Seedance 2.5 工作流中使用。在文本到视频、 每个草稿工作流都分两个阶段运行: 1. **创建草稿**:使用 **Seedance 2.5 Draft** 模型运行 Seedance 2.5 节点,获取 480p 预览及其 `draft_task_id`,**Save Text** 节点会将该 ID 写入输出文件夹中的 `.txt` 文件。可以用不同的提示词或输入重新运行,直到草稿看起来满意为止。 -2. **渲染成片**:从该 `.txt` 文件中复制 `draft_task_id`,填入 **ByteDance Seedance 2.5 Draft to Final Video** 节点上的 `draft_task_id` 字段并运行。该节点会渲染 1080p 成片,并复用创建草稿时的各项内容,包括提示词、输入、时长、宽高比和音频设置,因此无需重新输入任何内容。 +2. **渲染成片**:从该 `.txt` 文件中复制 `draft_task_id`,填入 **ByteDance Seedance 2.5 Draft to Final Video** 节点上的 `draft_task_id` 字段,将 `resolution` 设为 1080p 或 4K,然后运行。该节点会渲染成片,并复用创建草稿时的各项内容,包括提示词、输入、时长、宽高比和音频设置,因此无需重新输入任何内容。 每个模板都将这两个阶段放在单独的节点分组中,分别命名为 **Stage 1: Create and preview the draft video** 和 **Stage 2: Render the final video from the draft task ID**。对草稿满意后,请让 Stage 1 保持绕过状态,因为再次运行草稿节点会创建一个新草稿,而不是复用你已审阅的那个。 diff --git a/zh/tutorials/partner-nodes/bytedance/seedance-2-5.mdx b/zh/tutorials/partner-nodes/bytedance/seedance-2-5.mdx index 1abd64513a..22e065c1ba 100644 --- a/zh/tutorials/partner-nodes/bytedance/seedance-2-5.mdx +++ b/zh/tutorials/partner-nodes/bytedance/seedance-2-5.mdx @@ -2,13 +2,13 @@ title: "Seedance 2.5:带音频的 AI 视频生成" description: "使用 ComfyUI 中的 Seedance 2.5 生成 30 秒视频,支持同步音频、多模态参考控制与精确编辑" sidebarTitle: "Seedance 2.5" -translationSourceHash: 4dcc9b5c +translationSourceHash: 66330855 translationFrom: tutorials/partner-nodes/bytedance/seedance-2-5.mdx translationBlockHashes: "_intro": b6589680 "What Seedance 2.5 is good at": 5a91b792 "Example outputs": 23304d99 - "Available workflows": c27b5286 + "Available workflows": cf21df8c "Get started": 30f9b70a --- @@ -46,7 +46,7 @@ Seedance 2.5 是字节跳动的新一代视频创作模型,于 2026 年 7 月 ### 文生视频(T2V) -根据文本提示生成一段 30 秒的电影感视频,并带有同步的声音和音乐。该节点支持 480p、720p 和 1080p 输出分辨率。 +根据文本提示生成一段 30 秒的电影感视频,并带有同步的声音和音乐。该节点支持 480p、720p 和 1080p 输出分辨率,企业账号还可输出 4K。 @@ -183,9 +183,9 @@ Seedance 2.5 是字节跳动的新一代视频创作模型,于 2026 年 7 月 -### 草稿模式(先 480p 预览,再 1080p 成片) +### 草稿模式(先 480p 预览,再 1080p 或 4K 成片) -草稿模式会先渲染一段低成本的 480p 预览,再基于该草稿渲染出 1080p 成片。它适用于文生视频、首尾帧视频和参考生视频工作流。模板、示例输入和工作流技巧请参阅 [Seedance 2.5 草稿模式](/zh/tutorials/partner-nodes/bytedance/seedance-2-5-draft)。 +草稿模式会先渲染一段低成本的 480p 预览,再基于该草稿渲染出 1080p 或 4K 成片。它适用于文生视频、首尾帧视频和参考生视频工作流。模板、示例输入和工作流技巧请参阅 [Seedance 2.5 草稿模式](/zh/tutorials/partner-nodes/bytedance/seedance-2-5-draft)。 ## 开始使用 diff --git a/zh/tutorials/partner-nodes/pricing.mdx b/zh/tutorials/partner-nodes/pricing.mdx index f2774b920d..7d3cbeabb3 100644 --- a/zh/tutorials/partner-nodes/pricing.mdx +++ b/zh/tutorials/partner-nodes/pricing.mdx @@ -3,7 +3,7 @@ title: "Partner Node 定价" description: "各提供商(包括 Anthropic、OpenAI、Kling 和 Luma)的 Partner Node 积分定价,按模型、Token 类型和生成类型细分。" sidebarTitle: "定价" mode: "wide" -translationSourceHash: a08658db +translationSourceHash: 1091c8c7 translationFrom: tutorials/partner-nodes/pricing.mdx translationBlockHashes: "_intro": 1ec20604 @@ -11,7 +11,7 @@ translationBlockHashes: "Beeble": d1b19f61 "BFL": 7ae11dc6 "Bria": 94080005 - "ByteDance": d2af4769 + "ByteDance": e1a9d40e "Clarity AI": 3d568b52 "ElevenLabs": cadb48cc "FishAudio": c957b7da @@ -262,9 +262,13 @@ Seedream 5.0 Pro 使用基于分辨率的计费方式:**1K**(≤2.61 百 | Seedance 2.5 | 480p, 720p | 是 | 1.931 | | Seedance 2.5 | 1080p | 否 | 3.530 | | Seedance 2.5 | 1080p | 是 | 2.112 | +| Seedance 2.5 | 4K | 否 | 1.418 | +| Seedance 2.5 | 4K | 是 | 0.875 | **视频输入**表示已连接参考视频。文本、图片和音频参考均视为无视频输入。输入包含视频时适用最低 Token 消耗限制:若预计消耗低于最低值,按最低消耗计费。 +4K 输出仅企业账号可用。相同时长下,4K 单帧的 Token 数量约为 1080p 的四倍,因此每 1K Token 单价更低,但总费用更高。 + 参考视频至少需要 2 秒,输入总计不超过 30 秒。16:9 输出(24 fps)的预计费用: | 分辨率 | 5 秒输出,无视频输入 | 5 秒输出 + 2–30 秒视频输入 |