From 463e1b4f71ea9ebeb8204f03f0ba96360f0f2782 Mon Sep 17 00:00:00 2001 From: yajing Date: Tue, 18 Aug 2026 16:53:16 +0800 Subject: [PATCH 01/17] docs: add AI Credits guide --- docs.json | 8 +- en/cloud/use-dify/workspace/ai-credits.mdx | 96 +++++++++++++++++++ .../use-dify/workspace/model-providers.mdx | 32 +++---- 3 files changed, 116 insertions(+), 20 deletions(-) create mode 100644 en/cloud/use-dify/workspace/ai-credits.mdx diff --git a/docs.json b/docs.json index 9e944edcf..2b082ac7a 100644 --- a/docs.json +++ b/docs.json @@ -225,7 +225,13 @@ "expanded": false, "pages": [ "en/cloud/use-dify/workspace/plugins", - "en/cloud/use-dify/workspace/model-providers", + { + "group": "Model Providers", + "root": "en/cloud/use-dify/workspace/model-providers", + "pages": [ + "en/cloud/use-dify/workspace/ai-credits" + ] + }, "en/cloud/use-dify/workspace/tools", { "group": "Custom Endpoint", diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx new file mode 100644 index 000000000..8921159f0 --- /dev/null +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -0,0 +1,96 @@ +--- +title: AI Credits +description: Understand how Dify deducts AI Credits, compare model costs, and choose between credits and an API Key +--- + +Your plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. All apps in a workspace share the same AI Credits. + +## Calculate AI Credit Usage + +Your workspace may use AI Credits whenever a supported model runs. Common cases include: + +- **Running an app with an LLM**, including models configured directly in Chatbot, Text Generator, and Agent apps or in LLM, Question Classifier, Parameter Extractor, and Agent nodes within Workflows and Chatflows. +- **Using AI-assisted generation in the app editor** to create or revise prompts, code, structured output configurations, or Workflow and Chatflow drafts. +- **Indexing or searching a knowledge base**, which can involve an Embedding Model for indexing or vector search, an LLM when generating Q&A chunks from source documents, and a Rerank Model when reranking is enabled. +- **Transcribing voice input or generating audio output in an app** with a Speech-to-Text or Text-to-Speech Model. + +Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. For example, if a Workflow runs a 5-credit LLM twice and its Knowledge Retrieval node uses a 1-credit Rerank Model once, the run consumes 11 AI Credits. + +Your remaining balance must cover the full cost of each model use. A 5-credit model cannot run when only 4 AI Credits remain. + +## Compare Model Costs + +The following costs apply each time a model runs: + + + + + - **25 credits**: `gpt-5.6-sol`, `gpt-5.5`. + - **20 credits**: `gpt-4`. + - **15 credits**: `gpt-5.6-terra`, `gpt-5.4`. + - **10 credits**: `gpt-5.2`, `gpt-4o`, `gpt-4o-transcribe-diarize`, `gpt-4o-transcribe`, `whisper-1`, `gpt-4o-mini-tts-2025-12-15`, `tts-1-hd`, `tts-1`. + - **5 credits**: `gpt-5.6-luna`, `gpt-5.4-mini`, `gpt-5`, `gpt-5.1`, `gpt-4.1`, `o3-mini`, `o4-mini`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-mini-transcribe`, `text-embedding-3-small`, `text-embedding-3-large`, `text-embedding-ada-002`. + - **1 credit**: `gpt-5-mini`, `gpt-4.1-mini`, `gpt-4o-mini`, `gpt-3.5-turbo`. + + + + + - **20 credits**: `Claude Opus 4.7`, `Claude Opus 4.6`, `Claude Opus 4.5`. + - **10 credits**: `Claude Sonnet 4.6`, `Claude Sonnet 4.5`. + - **5 credits**: `Claude Haiku 4.5`. + + + + + - **10 credits**: `Gemini 3.5 Flash`, `Gemini 3.1 Pro-Preview`. + - **5 credits**: `gemini-3.7-flash`, `gemini-embedding-001`, `gemini-embedding-2`. + - **3 credits**: `Gemini 3 Flash-Preview`. + - **1 credit**: `Gemini 3.5 Flash-Lite`, `Gemini 3.1 Flash-Lite-Preview`, `Gemini 3.1 Flash-Lite`, `Gemini 2.5 Flash`, `Gemini 2.0 Flash`, `Gemini 2.0 Flash-Lite`. + + + + + - **20 credits**: `qwen3-tts-flash`. + - **10 credits**: `qwen3.7-max`, `qwen3.7-max-preview`, `qwen3.6-max-preview`, `qwen3-max`, `qwen3-max-preview`, `qwen-max`, `qwen3.6-plus`, `qwen-plus`, `qwen-plus-latest`, `qwen3.5-omni-plus`, `qvq-max`. + - **5 credits**: `qwen3.5-plus`, `qwen3.6-flash`, `qwen-flash`, `qwq-plus`, `qwen3.5-omni-flash`, `qwen3-vl-plus`, `qwen-vl-max`, `text-embedding-v3`, `text-embedding-v4`. + - **1 credit**: `qwen3.7-plus`, `qwen3.5-flash`, `qwen-turbo`, `qwen-long-latest`, `qwen-long-2025-01-25`, `qwen3-vl-flash`, `qwen-vl-plus`, `qwen3-rerank`. + + + + + +To use a model that is not listed above, configure your own API Key for its provider. + + +## Check Your AI Credits + +Go to **Integrations** > **Model Provider** to see how many AI Credits your workspace has used, the total, and the next reset date. If the credits have run out, the page also shows the date they ran out. + +Dify Cloud plans include the following AI Credits: + +| Plan | Included AI Credits | Renewal | +|:-----|:--------------------|:--------| +| Sandbox | 200 | One-time allocation | +| Professional | 5,000 | Monthly | +| Team | 10,000 | Monthly | + + +When you upgrade from Sandbox to Professional or Team, AI Credits are deducted from your new plan first. Any unused Sandbox credits become available after the new plan's credits run out. + + +## Set Usage Priority + +In **Integrations** > **Model Provider**, open the credential menu for a provider that supports AI Credits and has your API Key configured, then set **Usage Priority**. You set Usage Priority once for each provider, and the setting applies across the workspace. + +| Usage Priority | Used First | Switches To | +|:---------------|:-----------|:---------| +| **AI Credits** | Your workspace's AI Credits | Your API Key after all AI Credits run out | +| **API Key** | Your API Key | AI Credits if no API Key remains configured | + +Requests made with **API Key** use your provider account and do not deduct AI Credits. + +Adding the first API Key for a provider automatically sets **Usage Priority** to **API Key**. Change it to **AI Credits** if you want to use your workspace's credits first. + + +If a request using your API Key fails, it is not retried with AI Credits. Fix the API Key or provider issue, or change **Usage Priority** to **AI Credits** before retrying. + diff --git a/en/cloud/use-dify/workspace/model-providers.mdx b/en/cloud/use-dify/workspace/model-providers.mdx index 0ed8761ec..1f8c070b3 100644 --- a/en/cloud/use-dify/workspace/model-providers.mdx +++ b/en/cloud/use-dify/workspace/model-providers.mdx @@ -1,43 +1,37 @@ --- title: Model Providers -description: Use models with your AI credits, or connect your own provider accounts and keys +description: Use models with your AI Credits, or connect your own provider accounts and API Keys --- -Every AI app you build runs on a model. Dify Cloud comes with popular models, and lets you connect your own provider accounts when you need more. +Models power many of the apps and features you build in Dify. Dify Cloud comes with popular models and lets you connect your own provider accounts when you need more. -Models are shared across the workspace, so your whole team works with the same ones. +Configure a model provider once, then use its models across apps in the workspace. -Only the workspace owner and admins can manage providers. Any key you add works across the whole workspace and bills to your own account with that provider. +Only the workspace owner and admins can manage providers. Any API Key you add works across the whole workspace and bills to your own account with that provider. -## Use Models with Your AI Credits +## Use Models with AI Credits -Dify Cloud gives you AI credits to use models from a set of popular providers, so you don't need your own API key to start. - -Install one of these providers from **Integrations** > **Model Provider** (or the [Marketplace](https://marketplace.dify.ai/)), and its models are ready to use. Each AI response consumes your AI credits. - -An AI response is a single model call (one input and one output), and it counts as one response no matter how many tokens it uses. The number of credits a response costs depends on the model, with larger models costing more than smaller ones. - -See the [Dify pricing page](https://dify.ai/pricing) for which providers AI credits support and how many credits each model costs. +Dify Cloud includes AI Credits, so you can start using supported models without adding your own API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. ## Use Your Own Account with a Supported Provider -Add your own API key to a supported provider when you want higher rate limits or billing through your own provider account. +Add your own API Key to a supported provider when you want higher rate limits or billing through your own provider account. 1. In **Integrations** > **Model Provider**, install the provider if it isn't already. -2. Click **Setup** on its card, then enter your API key and any other required details. Dify checks the key before making the provider available. +2. Click **Setup** on its card, then enter your API Key and any other required details. Dify checks the API Key before making the provider available. -Your key and AI credits can coexist. Set **Usage Priority** on the provider card to control which one Dify draws from first and falls back to the other, so a key you add can simply take over once your credits run out. +Your API Key and AI Credits can coexist. See [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one requests use first. -## Connect a Provider AI Credits Don't Support +## Connect a Provider Not Supported by AI Credits -To use a provider that AI credits don't support, install it first and run it on your own account. AI credits don't apply to these providers. +To use a provider that AI Credits do not support, install it and add your own API Key. AI Credits do not apply to these providers. 1. In **Integrations** > **Model Provider**, browse **Install model providers**, or open the [Marketplace](https://marketplace.dify.ai/) for the full list. -2. Install the provider, click **Setup**, and enter your API key along with any other required details. +2. Install the provider, click **Setup**, and enter your API Key along with any other required details. ## Add a Custom Model @@ -115,4 +109,4 @@ Use load balancing to spread requests across several keys for the same model, so Switch back to a single key anytime; your load balancing setup is kept for later use. - \ No newline at end of file + From d70c0e41f23c817ce57ae8a2cf071efda5ec20d9 Mon Sep 17 00:00:00 2001 From: yajing Date: Tue, 18 Aug 2026 17:11:02 +0800 Subject: [PATCH 02/17] docs: refine AI Credits introduction --- en/cloud/use-dify/workspace/ai-credits.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 8921159f0..02d098565 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -1,9 +1,9 @@ --- title: AI Credits -description: Understand how Dify deducts AI Credits, compare model costs, and choose between credits and an API Key +description: Learn how your workspace uses AI Credits, what different models cost, and when requests use your own API Keys --- -Your plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. All apps in a workspace share the same AI Credits. +Each Dify Cloud plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. All apps in a workspace share the same AI Credits. ## Calculate AI Credit Usage From bfbf433c0a85a8140a0f4b520e860aa4ff7f3e22 Mon Sep 17 00:00:00 2001 From: yajing Date: Tue, 18 Aug 2026 17:13:27 +0800 Subject: [PATCH 03/17] docs: clarify AI Credit renewal status --- en/cloud/use-dify/workspace/ai-credits.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 02d098565..716585d0a 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -64,7 +64,7 @@ To use a model that is not listed above, configure your own API Key for its prov ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to see how many AI Credits your workspace has used, the total, and the next reset date. If the credits have run out, the page also shows the date they ran out. +Go to **Integrations** > **Model Provider** to check your workspace's AI Credits, including the amount used, the total, and the next renewal date. If no credits remain, the page also shows the date they reached zero. Dify Cloud plans include the following AI Credits: From 36ed0699ccbb0c559fc59505734b459e58f93d03 Mon Sep 17 00:00:00 2001 From: yajing Date: Tue, 18 Aug 2026 17:26:38 +0800 Subject: [PATCH 04/17] docs: refine AI Credit guidance --- en/cloud/use-dify/workspace/ai-credits.mdx | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 716585d0a..8e54663b2 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -1,11 +1,13 @@ --- title: AI Credits -description: Learn how your workspace uses AI Credits, what different models cost, and when requests use your own API Keys +description: Learn how your workspace uses AI Credits, what different models cost, and how to choose whether AI Credits or your own API Key is used first --- -Each Dify Cloud plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. All apps in a workspace share the same AI Credits. +Each Dify Cloud plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. -## Calculate AI Credit Usage +All apps in a workspace share the same AI Credits. + +## Understanding AI Credit Usage Your workspace may use AI Credits whenever a supported model runs. Common cases include: @@ -14,7 +16,9 @@ Your workspace may use AI Credits whenever a supported model runs. Common cases - **Indexing or searching a knowledge base**, which can involve an Embedding Model for indexing or vector search, an LLM when generating Q&A chunks from source documents, and a Rerank Model when reranking is enabled. - **Transcribing voice input or generating audio output in an app** with a Speech-to-Text or Text-to-Speech Model. -Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. For example, if a Workflow runs a 5-credit LLM twice and its Knowledge Retrieval node uses a 1-credit Rerank Model once, the run consumes 11 AI Credits. +Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. + +For example, if a Workflow runs a 5-credit LLM twice and its Knowledge Retrieval node uses a 1-credit Rerank Model once, the run consumes 11 AI Credits. Your remaining balance must cover the full cost of each model use. A 5-credit model cannot run when only 4 AI Credits remain. @@ -64,7 +68,7 @@ To use a model that is not listed above, configure your own API Key for its prov ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to check your workspace's AI Credits, including the amount used, the total, and the next renewal date. If no credits remain, the page also shows the date they reached zero. +Go to **Integrations** > **Model Provider** to check your workspace's AI Credits, including the amount used, the total, and the next renewal date. If your workspace has no AI Credits left, you can also see the date they ran out. Dify Cloud plans include the following AI Credits: @@ -75,11 +79,13 @@ Dify Cloud plans include the following AI Credits: | Team | 10,000 | Monthly | -When you upgrade from Sandbox to Professional or Team, AI Credits are deducted from your new plan first. Any unused Sandbox credits become available after the new plan's credits run out. +After you upgrade to Professional or Team, your new plan's AI Credits are used before any remaining Sandbox credits. ## Set Usage Priority +Usage Priority determines whether a provider uses AI Credits or your API Key first. Requests that use your API Key are billed to your provider account and do not deduct AI Credits. + In **Integrations** > **Model Provider**, open the credential menu for a provider that supports AI Credits and has your API Key configured, then set **Usage Priority**. You set Usage Priority once for each provider, and the setting applies across the workspace. | Usage Priority | Used First | Switches To | @@ -87,8 +93,6 @@ In **Integrations** > **Model Provider**, open the credential menu for a provide | **AI Credits** | Your workspace's AI Credits | Your API Key after all AI Credits run out | | **API Key** | Your API Key | AI Credits if no API Key remains configured | -Requests made with **API Key** use your provider account and do not deduct AI Credits. - Adding the first API Key for a provider automatically sets **Usage Priority** to **API Key**. Change it to **AI Credits** if you want to use your workspace's credits first. From 26258983ea4369937fb81de840004d7cd506e73d Mon Sep 17 00:00:00 2001 From: yajing Date: Tue, 18 Aug 2026 18:02:26 +0800 Subject: [PATCH 05/17] docs: refine ai credits guidance --- en/cloud/use-dify/workspace/ai-credits.mdx | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 8e54663b2..bf1f2b531 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -1,6 +1,6 @@ --- title: AI Credits -description: Learn how your workspace uses AI Credits, what different models cost, and how to choose whether AI Credits or your own API Key is used first +description: Learn how AI Credits work in your workspace, from model costs to usage priority --- Each Dify Cloud plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. @@ -20,7 +20,7 @@ Each model available through AI Credits has a fixed cost, regardless of input or For example, if a Workflow runs a 5-credit LLM twice and its Knowledge Retrieval node uses a 1-credit Rerank Model once, the run consumes 11 AI Credits. -Your remaining balance must cover the full cost of each model use. A 5-credit model cannot run when only 4 AI Credits remain. +Each time a model runs, your workspace must have enough AI Credits to cover its full cost. A 5-credit model cannot run when only 4 credits remain. ## Compare Model Costs @@ -70,7 +70,7 @@ To use a model that is not listed above, configure your own API Key for its prov Go to **Integrations** > **Model Provider** to check your workspace's AI Credits, including the amount used, the total, and the next renewal date. If your workspace has no AI Credits left, you can also see the date they ran out. -Dify Cloud plans include the following AI Credits: +The number of included AI Credits depends on your plan: | Plan | Included AI Credits | Renewal | |:-----|:--------------------|:--------| @@ -84,14 +84,12 @@ After you upgrade to Professional or Team, your new plan's AI Credits are used b ## Set Usage Priority -Usage Priority determines whether a provider uses AI Credits or your API Key first. Requests that use your API Key are billed to your provider account and do not deduct AI Credits. +For a provider that supports AI Credits and has your API Key configured, Usage Priority determines which one is used first. Requests that use your API Key are billed to your provider account and do not deduct AI Credits. -In **Integrations** > **Model Provider**, open the credential menu for a provider that supports AI Credits and has your API Key configured, then set **Usage Priority**. You set Usage Priority once for each provider, and the setting applies across the workspace. +After adding an API Key, go to **Integrations** > **Model Provider** and click **Configure** on the provider card to set **Usage Priority**. Each provider has its own setting, which applies to every app in the workspace that uses that provider. -| Usage Priority | Used First | Switches To | -|:---------------|:-----------|:---------| -| **AI Credits** | Your workspace's AI Credits | Your API Key after all AI Credits run out | -| **API Key** | Your API Key | AI Credits if no API Key remains configured | +- Select **AI Credits** to use AI Credits first. After they run out, your API Key is used. +- Select **API Key** to use your API Key first. If no API Key remains configured, available AI Credits are used. Adding the first API Key for a provider automatically sets **Usage Priority** to **API Key**. Change it to **AI Credits** if you want to use your workspace's credits first. From 7287bd15fb38b1d6c7bfefed7cac6c91665f9fb5 Mon Sep 17 00:00:00 2001 From: yajing Date: Tue, 18 Aug 2026 18:30:45 +0800 Subject: [PATCH 06/17] docs: simplify usage priority guidance --- en/cloud/use-dify/workspace/ai-credits.mdx | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index bf1f2b531..d1c5ab332 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -84,15 +84,13 @@ After you upgrade to Professional or Team, your new plan's AI Credits are used b ## Set Usage Priority -For a provider that supports AI Credits and has your API Key configured, Usage Priority determines which one is used first. Requests that use your API Key are billed to your provider account and do not deduct AI Credits. +For a provider that supports AI Credits, the first API Key you add is used before AI Credits by default. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: -After adding an API Key, go to **Integrations** > **Model Provider** and click **Configure** on the provider card to set **Usage Priority**. Each provider has its own setting, which applies to every app in the workspace that uses that provider. +- Select **AI Credits** to use them first. After they run out, your API Key is used. +- Select **API Key** to use your provider account instead of AI Credits. If no API Key can be used, available AI Credits are used instead. -- Select **AI Credits** to use AI Credits first. After they run out, your API Key is used. -- Select **API Key** to use your API Key first. If no API Key remains configured, available AI Credits are used. - -Adding the first API Key for a provider automatically sets **Usage Priority** to **API Key**. Change it to **AI Credits** if you want to use your workspace's credits first. +The setting applies to that provider across the workspace. -If a request using your API Key fails, it is not retried with AI Credits. Fix the API Key or provider issue, or change **Usage Priority** to **AI Credits** before retrying. +A failed API Key request is not retried with AI Credits. Fix the API Key or provider issue, or select **AI Credits** before retrying. From 77f958c5bb9f1e3b300eb44f0f57c518cb10972c Mon Sep 17 00:00:00 2001 From: yajing Date: Tue, 18 Aug 2026 23:48:52 +0800 Subject: [PATCH 07/17] docs: refine AI credit guidance --- en/cloud/use-dify/workspace/ai-credits.mdx | 27 ++++++---- .../use-dify/workspace/model-providers.mdx | 52 +++++++++---------- 2 files changed, 42 insertions(+), 37 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index d1c5ab332..b589399aa 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -11,10 +11,17 @@ All apps in a workspace share the same AI Credits. Your workspace may use AI Credits whenever a supported model runs. Common cases include: -- **Running an app with an LLM**, including models configured directly in Chatbot, Text Generator, and Agent apps or in LLM, Question Classifier, Parameter Extractor, and Agent nodes within Workflows and Chatflows. -- **Using AI-assisted generation in the app editor** to create or revise prompts, code, structured output configurations, or Workflow and Chatflow drafts. -- **Indexing or searching a knowledge base**, which can involve an Embedding Model for indexing or vector search, an LLM when generating Q&A chunks from source documents, and a Rerank Model when reranking is enabled. -- **Transcribing voice input or generating audio output in an app** with a Speech-to-Text or Text-to-Speech Model. +- **Running an app with an LLM** + - Chatbot, Text Generator, and Agent apps run their configured LLM. + - Workflows and Chatflows can run LLMs through LLM, Question Classifier, Parameter Extractor, and Agent nodes. +- **Using AI-assisted generation in the app editor** + - An LLM can create or revise prompts, code, structured output configurations, or Workflow and Chatflow drafts. +- **Indexing or searching a knowledge base** + - An Embedding Model can be used for indexing or vector search. + - An LLM can generate Q&A chunks from source documents. + - When reranking is enabled, a Rerank Model can reorder search results. +- **Transcribing voice input or generating audio output in an app** + - A Speech-to-Text Model transcribes voice input, while a Text-to-Speech Model generates audio output. Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. @@ -63,12 +70,12 @@ The following costs apply each time a model runs: -To use a model that is not listed above, configure your own API Key for its provider. +To use a model that is not listed above, [add your own API Key for its provider](/en/cloud/use-dify/workspace/model-providers#use-your-own-account-with-a-provider-that-supports-ai-credits). ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to check your workspace's AI Credits, including the amount used, the total, and the next renewal date. If your workspace has no AI Credits left, you can also see the date they ran out. +Go to **Integrations** > **Model Provider** to check your workspace's AI Credit usage and total. You can also find the next renewal date for Professional and Team plans or the date your credits ran out. The number of included AI Credits depends on your plan: @@ -84,13 +91,11 @@ After you upgrade to Professional or Team, your new plan's AI Credits are used b ## Set Usage Priority -For a provider that supports AI Credits, the first API Key you add is used before AI Credits by default. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: +For a provider that supports AI Credits, the first API Key you add is automatically used before AI Credits. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: -- Select **AI Credits** to use them first. After they run out, your API Key is used. +- Select **AI Credits** to use them first. After they run out, an available API Key is used. - Select **API Key** to use your provider account instead of AI Credits. If no API Key can be used, available AI Credits are used instead. -The setting applies to that provider across the workspace. - -A failed API Key request is not retried with AI Credits. Fix the API Key or provider issue, or select **AI Credits** before retrying. +If a request sent with your API Key fails, it is not retried with AI Credits. Fix the API Key or provider issue, or select **AI Credits** before retrying. diff --git a/en/cloud/use-dify/workspace/model-providers.mdx b/en/cloud/use-dify/workspace/model-providers.mdx index 1f8c070b3..9d0a20fa3 100644 --- a/en/cloud/use-dify/workspace/model-providers.mdx +++ b/en/cloud/use-dify/workspace/model-providers.mdx @@ -3,27 +3,27 @@ title: Model Providers description: Use models with your AI Credits, or connect your own provider accounts and API Keys --- -Models power many of the apps and features you build in Dify. Dify Cloud comes with popular models and lets you connect your own provider accounts when you need more. +Models power many of the apps and features you build in Dify. Dify Cloud provides two ways to access them: use supported models with the AI Credits included in your plan, or add your own API Key to connect a provider account. Configure a model provider once, then use its models across apps in the workspace. - -Only the workspace owner and admins can manage providers. Any API Key you add works across the whole workspace and bills to your own account with that provider. - - ## Use Models with AI Credits -Dify Cloud includes AI Credits, so you can start using supported models without adding your own API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. +With AI Credits, you can start using supported models without adding API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. + +## Use Your Own Account with a Provider That Supports AI Credits -## Use Your Own Account with a Supported Provider +Add your own API Key to use the model access, rate limits, and billing available through your provider account. -Add your own API Key to a supported provider when you want higher rate limits or billing through your own provider account. + +Only the workspace owner and admins can add or manage provider API Keys. + 1. In **Integrations** > **Model Provider**, install the provider if it isn't already. -2. Click **Setup** on its card, then enter your API Key and any other required details. Dify checks the API Key before making the provider available. +2. Click **Add API Key** on its card, then enter your API Key and any other required details. Dify validates the API Key before it can be used. -Your API Key and AI Credits can coexist. See [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one requests use first. +Your API Key and AI Credits can coexist. Requests using your API Key go through your provider account and do not deduct AI Credits. See [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one is used first. ## Connect a Provider Not Supported by AI Credits @@ -31,7 +31,7 @@ To use a provider that AI Credits do not support, install it and add your own AP 1. In **Integrations** > **Model Provider**, browse **Install model providers**, or open the [Marketplace](https://marketplace.dify.ai/) for the full list. -2. Install the provider, click **Setup**, and enter your API Key along with any other required details. +2. Install the provider, click **Add API Key**, and enter your API Key along with any other required details. ## Add a Custom Model @@ -40,18 +40,18 @@ A provider's models are ready as soon as you connect it, so you only add one by Click **Add Model** on the provider's card, then give the model a name and its credentials. Providers that serve only a fixed set of models don't offer this option. -If the model you add matches the name and type of an existing one, Dify attaches the new key to that model instead of creating a duplicate. +If the model you add matches the name and type of an existing one, Dify attaches the new API Key to that model instead of creating a duplicate. -## Manage Your Keys +## Manage API Keys -Add more than one key to a provider when you want to keep development and production apart or spread usage across several accounts. +Add more than one API Key to a provider when you need to switch between provider accounts or configure load balancing for a model. -- For a provider's own models, click **Configure** on its card to manage the keys they share. +- For the models listed on a provider card, click **Configure** on the card to manage the API Keys they share. ![Manage Credentials for a Provider's Models](/images/use-dify/workspace/predefined-model-credential.png) -- For a custom model you added, click **Configure** on that model to manage its own keys. +- For a custom model you added, click **Configure** on that model to manage its own API Keys. ![Custom Model Configuration](/images/use-dify/workspace/custom-model-config.png) @@ -62,16 +62,16 @@ Add more than one key to a provider when you want to keep development and produc - A custom model relies on its own keys. Deleting its only key removes the model too. + A custom model relies on its own API Keys. Deleting its only API Key removes the model too. - To view every custom model's keys in one place, click **Manage Credentials** on the provider card. + To view every custom model's API Keys in one place, click **Manage Credentials** on the provider card. ![Manage Credentials for All Custom Models](/images/use-dify/workspace/custom-model-credential-list.png) - Keys stay here even after the custom model is removed, so you can re-add the model later without re-entering its keys. + API Keys stay here even after the custom model is removed, so you can re-add the model later without re-entering them. ![Re-add a Removed Model](/images/use-dify/workspace/removed-custom-model-re-add.png) @@ -87,26 +87,26 @@ Apps and nodes that don't pick a model fall back to your workspace defaults. Cli - **Speech-to-Text Model**: turns audio into text. - **Text-to-Speech Model**: turns text into audio. -## Spread Requests Across Keys with Load Balancing Professional Team +## Spread Requests Across API Keys with Load Balancing Professional Team -When one key handles many requests at once, it can hit the provider's rate limit and start failing. +When one API Key handles many requests at once, it can hit the provider's rate limit and start failing. -Use load balancing to spread requests across several keys for the same model, so no single key becomes the bottleneck. Dify rotates through the keys in turn and rests any that hits its limit for a minute before trying it again. +Use load balancing to spread requests across several API Keys for the same model, so no single API Key becomes the bottleneck. Dify rotates through the API Keys in turn and rests any that hits its limit for a minute before trying it again. 1. Find the model in the list, click **Configure**, and select **Load Balancing**. -2. Click **Add credential** to add keys to the pool. +2. Click **Add credential** to add API Keys to the pool. - Add a higher-quota or faster key more than once to send more of the traffic its way. + Add a higher-quota or faster API Key more than once to send more of the traffic its way. ![Add Credentials for Load Balancing](/images/use-dify/workspace/add-load-balancing-credential.png) -3. Turn on at least two keys, then click **Save**. A model using load balancing shows a marker in the list. +3. Turn on at least two API Keys, then click **Save**. A model using load balancing shows a marker in the list. -Switch back to a single key anytime; your load balancing setup is kept for later use. +Switch back to a single API Key anytime; your load balancing setup is kept for later use. From 66c1ce91258aa55889c89cadd7bd87c63fe50984 Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 00:05:10 +0800 Subject: [PATCH 08/17] docs: streamline model provider setup --- docs.json | 4 +- en/cloud/use-dify/workspace/ai-credits.mdx | 17 +++----- .../use-dify/workspace/model-providers.mdx | 42 +++++++++---------- 3 files changed, 26 insertions(+), 37 deletions(-) diff --git a/docs.json b/docs.json index 2b082ac7a..e697aa8f7 100644 --- a/docs.json +++ b/docs.json @@ -5134,11 +5134,11 @@ }, { "source": "/en/guides/model-configuration/load-balancing", - "destination": "/en/cloud/use-dify/workspace/model-providers#configure-model-load-balancing" + "destination": "/en/cloud/use-dify/workspace/model-providers#spread-requests-across-api-keys-with-load-balancing" }, { "source": "/guides/model-configuration/load-balancing", - "destination": "/en/cloud/use-dify/workspace/model-providers#configure-model-load-balancing" + "destination": "/en/cloud/use-dify/workspace/model-providers#spread-requests-across-api-keys-with-load-balancing" }, { "source": "/zh-hans/guides/model-configuration/load-balancing", diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index b589399aa..ca1eb1925 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -11,17 +11,10 @@ All apps in a workspace share the same AI Credits. Your workspace may use AI Credits whenever a supported model runs. Common cases include: -- **Running an app with an LLM** - - Chatbot, Text Generator, and Agent apps run their configured LLM. - - Workflows and Chatflows can run LLMs through LLM, Question Classifier, Parameter Extractor, and Agent nodes. -- **Using AI-assisted generation in the app editor** - - An LLM can create or revise prompts, code, structured output configurations, or Workflow and Chatflow drafts. -- **Indexing or searching a knowledge base** - - An Embedding Model can be used for indexing or vector search. - - An LLM can generate Q&A chunks from source documents. - - When reranking is enabled, a Rerank Model can reorder search results. -- **Transcribing voice input or generating audio output in an app** - - A Speech-to-Text Model transcribes voice input, while a Text-to-Speech Model generates audio output. +- **Running an app with an LLM**, including models configured directly in Chatbot, Text Generator, and Agent apps or in LLM, Question Classifier, Parameter Extractor, and Agent nodes within Workflows and Chatflows. +- **Using AI-assisted generation in the app editor** to create or revise prompts, code, structured output configurations, or Workflow and Chatflow drafts. +- **Indexing or searching a knowledge base**, which can involve an Embedding Model for indexing or vector search, an LLM when generating Q&A chunks from source documents, and a Rerank Model when reranking is enabled. +- **Transcribing voice input or generating audio output in an app** with a Speech-to-Text or Text-to-Speech Model. Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. @@ -70,7 +63,7 @@ The following costs apply each time a model runs: -To use a model that is not listed above, [add your own API Key for its provider](/en/cloud/use-dify/workspace/model-providers#use-your-own-account-with-a-provider-that-supports-ai-credits). +To use a model that is not listed above, [add your own API Key for its provider](/en/cloud/use-dify/workspace/model-providers#use-your-own-api-key). ## Check Your AI Credits diff --git a/en/cloud/use-dify/workspace/model-providers.mdx b/en/cloud/use-dify/workspace/model-providers.mdx index 9d0a20fa3..09b26628f 100644 --- a/en/cloud/use-dify/workspace/model-providers.mdx +++ b/en/cloud/use-dify/workspace/model-providers.mdx @@ -7,31 +7,27 @@ Models power many of the apps and features you build in Dify. Dify Cloud provide Configure a model provider once, then use its models across apps in the workspace. -## Use Models with AI Credits +## Use AI Credits -With AI Credits, you can start using supported models without adding API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. +With AI Credits, you can use supported models without adding API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. -## Use Your Own Account with a Provider That Supports AI Credits +## Use Your Own API Key -Add your own API Key to use the model access, rate limits, and billing available through your provider account. +Add your own API Key to access models through your provider account. Only the workspace owner and admins can add or manage provider API Keys. -1. In **Integrations** > **Model Provider**, install the provider if it isn't already. +1. Go to **Integrations** > **Model Provider**. -2. Click **Add API Key** on its card, then enter your API Key and any other required details. Dify validates the API Key before it can be used. +2. Find the provider. If it is not installed, select it under **Install model providers** or find it in the [Marketplace](https://marketplace.dify.ai/). -Your API Key and AI Credits can coexist. Requests using your API Key go through your provider account and do not deduct AI Credits. See [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one is used first. +3. Click **Add API Key**, then enter your API Key and any other required details. -## Connect a Provider Not Supported by AI Credits +If the provider supports AI Credits, both options remain available. Set [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one is used first. -To use a provider that AI Credits do not support, install it and add your own API Key. AI Credits do not apply to these providers. - -1. In **Integrations** > **Model Provider**, browse **Install model providers**, or open the [Marketplace](https://marketplace.dify.ai/) for the full list. - -2. Install the provider, click **Add API Key**, and enter your API Key along with any other required details. +Providers that do not support AI Credits use your API Key. Requests using it go through your provider account and do not deduct AI Credits. ## Add a Custom Model @@ -43,6 +39,16 @@ Click **Add Model** on the provider's card, then give the model a name and its c If the model you add matches the name and type of an existing one, Dify attaches the new API Key to that model instead of creating a duplicate. +## Choose the Models Apps Use by Default + +Apps and nodes that don't pick a model fall back to your workspace defaults. Click **Default Models** at the top-right corner to set one for each job: + +- **System Reasoning Model**: the default for general LLM tasks. +- **Embedding Model**: indexes and retrieves knowledge base content. +- **Rerank Model**: reorders retrieval results by relevance. +- **Speech-to-Text Model**: turns audio into text. +- **Text-to-Speech Model**: turns text into audio. + ## Manage API Keys Add more than one API Key to a provider when you need to switch between provider accounts or configure load balancing for a model. @@ -77,16 +83,6 @@ Add more than one API Key to a provider when you need to switch between provider ![Re-add a Removed Model](/images/use-dify/workspace/removed-custom-model-re-add.png) -## Choose the Models Apps Use by Default - -Apps and nodes that don't pick a model fall back to your workspace defaults. Click **Default Models** at the top-right corner to set one for each job: - -- **System Reasoning Model**: the default for general LLM tasks. -- **Embedding Model**: indexes and retrieves knowledge base content. -- **Rerank Model**: reorders retrieval results by relevance. -- **Speech-to-Text Model**: turns audio into text. -- **Text-to-Speech Model**: turns text into audio. - ## Spread Requests Across API Keys with Load Balancing Professional Team When one API Key handles many requests at once, it can hit the provider's rate limit and start failing. From 86b5c9aada9b0d7e04e587320ee3a10fd62f7d0a Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 00:15:31 +0800 Subject: [PATCH 09/17] docs: align AI credit section heading --- en/cloud/use-dify/workspace/ai-credits.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index ca1eb1925..b33cc0f65 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -7,7 +7,7 @@ Each Dify Cloud plan includes a set number of AI Credits for using supported mod All apps in a workspace share the same AI Credits. -## Understanding AI Credit Usage +## Understand AI Credit Usage Your workspace may use AI Credits whenever a supported model runs. Common cases include: From 4c2f865236428025f9bd03b91f99c194de960ae5 Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 14:14:09 +0800 Subject: [PATCH 10/17] docs: clarify AI credit usage --- en/cloud/use-dify/workspace/ai-credits.mdx | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index b33cc0f65..d933ea940 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -5,7 +5,7 @@ description: Learn how AI Credits work in your workspace, from model costs to us Each Dify Cloud plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. -All apps in a workspace share the same AI Credits. +All app runs draw from the workspace's shared AI Credits, including runs from the editor, published WebApps, the Service API, and triggers. ## Understand AI Credit Usage @@ -18,7 +18,7 @@ Your workspace may use AI Credits whenever a supported model runs. Common cases Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. -For example, if a Workflow runs a 5-credit LLM twice and its Knowledge Retrieval node uses a 1-credit Rerank Model once, the run consumes 11 AI Credits. +Suppose you index documents in a knowledge base, then test a Workflow that searches it with vector search and reranking enabled. If indexing runs a 5-credit Embedding Model three times, it uses 15 AI Credits. The Workflow test runs the Embedding Model once for the search, a 1-credit Rerank Model once, and a 5-credit LLM twice, using another 16 AI Credits. Each time a model runs, your workspace must have enough AI Credits to cover its full cost. A 5-credit model cannot run when only 4 credits remain. @@ -68,7 +68,7 @@ To use a model that is not listed above, [add your own API Key for its provider] ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to check your workspace's AI Credit usage and total. You can also find the next renewal date for Professional and Team plans or the date your credits ran out. +Go to **Integrations** > **Model Provider** to check your workspace's AI Credit usage and total. If no credits remain, the page shows the date they ran out. On Professional and Team plans, you can also check the next renewal date. The number of included AI Credits depends on your plan: @@ -79,15 +79,15 @@ The number of included AI Credits depends on your plan: | Team | 10,000 | Monthly | -After you upgrade to Professional or Team, your new plan's AI Credits are used before any remaining Sandbox credits. +After you upgrade to Professional or Team, your new plan's AI Credits are used before any remaining Sandbox credits. The two credit pools are not combined for a single model run. For example, if 4 plan credits remain, a model that costs 5 credits cannot use 1 Sandbox credit to cover the difference. ## Set Usage Priority For a provider that supports AI Credits, the first API Key you add is automatically used before AI Credits. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: -- Select **AI Credits** to use them first. After they run out, an available API Key is used. -- Select **API Key** to use your provider account instead of AI Credits. If no API Key can be used, available AI Credits are used instead. +- Select **AI Credits** to use AI Credits first, then an available API Key after the credits run out. +- Select **API Key** to use a configured API Key first, then available AI Credits if no API Key is usable. If a request sent with your API Key fails, it is not retried with AI Credits. Fix the API Key or provider issue, or select **AI Credits** before retrying. From b87230507f9239db91af36ce7f79af6ce57156dd Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 14:37:00 +0800 Subject: [PATCH 11/17] docs: simplify AI credit examples --- en/cloud/use-dify/workspace/ai-credits.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index d933ea940..09b77258b 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -18,7 +18,7 @@ Your workspace may use AI Credits whenever a supported model runs. Common cases Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. -Suppose you index documents in a knowledge base, then test a Workflow that searches it with vector search and reranking enabled. If indexing runs a 5-credit Embedding Model three times, it uses 15 AI Credits. The Workflow test runs the Embedding Model once for the search, a 1-credit Rerank Model once, and a 5-credit LLM twice, using another 16 AI Credits. +For example, you might index documents in a knowledge base and then test a Workflow that searches it with vector search. If indexing runs a 1-credit Embedding Model three times, it uses 3 AI Credits. The Workflow then runs the Embedding Model once and a 5-credit LLM twice, bringing the total to 14 AI Credits. Each time a model runs, your workspace must have enough AI Credits to cover its full cost. A 5-credit model cannot run when only 4 credits remain. @@ -79,7 +79,7 @@ The number of included AI Credits depends on your plan: | Team | 10,000 | Monthly | -After you upgrade to Professional or Team, your new plan's AI Credits are used before any remaining Sandbox credits. The two credit pools are not combined for a single model run. For example, if 4 plan credits remain, a model that costs 5 credits cannot use 1 Sandbox credit to cover the difference. +After you upgrade to Professional or Team, your new plan's AI Credits are used before any remaining Sandbox credits. The two credit pools are not combined for a single model run. ## Set Usage Priority From e9cecaee09c006d0f6cbc3d14490137f6e54a2a3 Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 14:39:42 +0800 Subject: [PATCH 12/17] docs: restore AI credit example wording --- en/cloud/use-dify/workspace/ai-credits.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 09b77258b..63f1d7154 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -18,7 +18,7 @@ Your workspace may use AI Credits whenever a supported model runs. Common cases Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. -For example, you might index documents in a knowledge base and then test a Workflow that searches it with vector search. If indexing runs a 1-credit Embedding Model three times, it uses 3 AI Credits. The Workflow then runs the Embedding Model once and a 5-credit LLM twice, bringing the total to 14 AI Credits. +Suppose you index documents in a knowledge base, then test a Workflow that searches it with vector search. Indexing runs a 1-credit Embedding Model three times and uses 3 AI Credits. The Workflow then runs the Embedding Model once and a 5-credit LLM twice, bringing the total to 14 AI Credits. Each time a model runs, your workspace must have enough AI Credits to cover its full cost. A 5-credit model cannot run when only 4 credits remain. From bf92adbdbace70a2611e52595cb76b7a36388ebe Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 15:54:10 +0800 Subject: [PATCH 13/17] docs: refine AI credit usage guidance --- en/cloud/use-dify/workspace/ai-credits.mdx | 32 +++++++++---------- .../use-dify/workspace/model-providers.mdx | 14 ++++---- tools/translate/termbase_i18n.md | 7 ++-- writing-guides/glossary.md | 10 +++--- 4 files changed, 31 insertions(+), 32 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 63f1d7154..0d252adc4 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -1,26 +1,24 @@ --- title: AI Credits -description: Learn how AI Credits work in your workspace, from model costs to usage priority +description: Learn how AI credits work in your workspace, from model costs to usage priority --- -Each Dify Cloud plan includes a set number of AI Credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. +Each Dify Cloud plan includes a set number of AI credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. -All app runs draw from the workspace's shared AI Credits, including runs from the editor, published WebApps, the Service API, and triggers. +App runs that use AI credits draw from the workspace's shared balance, including runs from the editor, published web apps, the Service API, and triggers. ## Understand AI Credit Usage -Your workspace may use AI Credits whenever a supported model runs. Common cases include: +Your workspace may use AI credits whenever a supported model runs. Common cases include: - **Running an app with an LLM**, including models configured directly in Chatbot, Text Generator, and Agent apps or in LLM, Question Classifier, Parameter Extractor, and Agent nodes within Workflows and Chatflows. - **Using AI-assisted generation in the app editor** to create or revise prompts, code, structured output configurations, or Workflow and Chatflow drafts. -- **Indexing or searching a knowledge base**, which can involve an Embedding Model for indexing or vector search, an LLM when generating Q&A chunks from source documents, and a Rerank Model when reranking is enabled. -- **Transcribing voice input or generating audio output in an app** with a Speech-to-Text or Text-to-Speech Model. +- **Indexing or searching a knowledge base**, which can involve an embedding model for indexing or vector search, an LLM when generating Q&A chunks from source documents, and a rerank model when reranking is enabled. +- **Transcribing voice input or generating audio output in an app** with a speech-to-text or text-to-speech model. -Each model available through AI Credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. +Each model available through AI credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. -Suppose you index documents in a knowledge base, then test a Workflow that searches it with vector search. Indexing runs a 1-credit Embedding Model three times and uses 3 AI Credits. The Workflow then runs the Embedding Model once and a 5-credit LLM twice, bringing the total to 14 AI Credits. - -Each time a model runs, your workspace must have enough AI Credits to cover its full cost. A 5-credit model cannot run when only 4 credits remain. +Suppose you index documents in a knowledge base, then test a Workflow that searches it with vector search. Indexing runs a 5-credit embedding model three times and uses 15 AI credits. The Workflow then runs the embedding model once and a 5-credit LLM twice, bringing the total to 30 AI credits. ## Compare Model Costs @@ -68,9 +66,9 @@ To use a model that is not listed above, [add your own API Key for its provider] ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to check your workspace's AI Credit usage and total. If no credits remain, the page shows the date they ran out. On Professional and Team plans, you can also check the next renewal date. +Go to **Integrations** > **Model Provider** to check how many AI credits your workspace has used out of the current total. If no credits remain, the page shows the date they ran out. On Professional and Team plans, you can also check the next renewal date. -The number of included AI Credits depends on your plan: +The number of included AI credits depends on your plan: | Plan | Included AI Credits | Renewal | |:-----|:--------------------|:--------| @@ -79,16 +77,16 @@ The number of included AI Credits depends on your plan: | Team | 10,000 | Monthly | -After you upgrade to Professional or Team, your new plan's AI Credits are used before any remaining Sandbox credits. The two credit pools are not combined for a single model run. +After upgrading to Professional or Team, your new plan's AI credits are used before any remaining Sandbox credits. The two balances cannot be combined for one model run. ## Set Usage Priority -For a provider that supports AI Credits, the first API Key you add is automatically used before AI Credits. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: +For a provider that supports AI credits, the first API Key you add is automatically used before AI credits. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: -- Select **AI Credits** to use AI Credits first, then an available API Key after the credits run out. -- Select **API Key** to use a configured API Key first, then available AI Credits if no API Key is usable. +- **AI credits first**: Requests use AI credits until they run out, then switch to the selected API Key. If the remaining balance cannot cover the model's full cost, change the priority to **API Key**. +- **API Key first**: Requests use the selected API Key. If it is unavailable when a request starts, the request uses available AI credits. -If a request sent with your API Key fails, it is not retried with AI Credits. Fix the API Key or provider issue, or select **AI Credits** before retrying. +If a request has already been sent with your API Key and fails, it is not retried with AI credits. Fix the API Key or provider issue, or change the priority to **AI credits** before retrying. diff --git a/en/cloud/use-dify/workspace/model-providers.mdx b/en/cloud/use-dify/workspace/model-providers.mdx index 09b26628f..40cc704b3 100644 --- a/en/cloud/use-dify/workspace/model-providers.mdx +++ b/en/cloud/use-dify/workspace/model-providers.mdx @@ -1,15 +1,15 @@ --- title: Model Providers -description: Use models with your AI Credits, or connect your own provider accounts and API Keys +description: Use models with your AI credits, or connect your own provider accounts and API Keys --- -Models power many of the apps and features you build in Dify. Dify Cloud provides two ways to access them: use supported models with the AI Credits included in your plan, or add your own API Key to connect a provider account. +Models power many of the apps and features you build in Dify. Dify Cloud provides two ways to access them: use supported models with the AI credits included in your plan, or add your own API Key to connect a provider account. -Configure a model provider once, then use its models across apps in the workspace. +Both options let apps across the workspace access the provider's available models. ## Use AI Credits -With AI Credits, you can use supported models without adding API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. +With AI credits, you can use supported models without adding API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. ## Use Your Own API Key @@ -25,9 +25,9 @@ Only the workspace owner and admins can add or manage provider API Keys. 3. Click **Add API Key**, then enter your API Key and any other required details. -If the provider supports AI Credits, both options remain available. Set [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one is used first. +If the provider supports AI credits, both options remain available. Set [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one is used first. -Providers that do not support AI Credits use your API Key. Requests using it go through your provider account and do not deduct AI Credits. +Requests made with your API Key go through your provider account and do not deduct AI credits. For providers that do not support AI credits, an API Key is the only option. ## Add a Custom Model @@ -89,7 +89,7 @@ When one API Key handles many requests at once, it can hit the provider's rate l Use load balancing to spread requests across several API Keys for the same model, so no single API Key becomes the bottleneck. Dify rotates through the API Keys in turn and rests any that hits its limit for a minute before trying it again. -1. Find the model in the list, click **Configure**, and select **Load Balancing**. +1. Find the model in the list, click **Configure**, and select **Load balancing**. 2. Click **Add credential** to add API Keys to the pool. diff --git a/tools/translate/termbase_i18n.md b/tools/translate/termbase_i18n.md index c529cddba..bc86a3efb 100644 --- a/tools/translate/termbase_i18n.md +++ b/tools/translate/termbase_i18n.md @@ -28,6 +28,7 @@ |:--------|:--------|:---------| | model | 模型 | モデル | | model provider | 模型供应商 | モデルプロバイダー | +| AI credits | AI 消息额度 | AI クレジット | | LLM (large language model) | 大语言模型 | 大規模言語モデル | | chat model | 对话模型 | チャットモデル | | completion model | 文本续写模型 | 補完モデル | @@ -37,8 +38,8 @@ | rerank model | 重排序模型 | リランクモデル | | reasoning model | 推理模型 | 推論モデル | | moderation model | 内容审核模型 | モデレーションモデル | -| TTS model | 文字转语音模型 | TTSモデル | -| Speech2Text model | 语音转文字模型 | 音声認識モデル | +| speech-to-text model | 语音转文字模型 | 音声認識モデル | +| text-to-speech model | 文字转语音模型 | TTSモデル | | token | token | token | | API token | API 令牌 | APIトークン | | prompt | 提示词 | プロンプト | @@ -218,7 +219,7 @@ | Rerank Model | Rerank 模型 | Rerank モデル | | Speech-to-Text Model | 语音转文本模型 | 音声-to-テキストモデル | | Text-to-Speech Model | 文本转语音模型 | テキスト-to-音声モデル | -| Load Balancing | 负载均衡 | 負荷分散 | +| Load balancing | 负载均衡 | 負荷分散 | | AI credits | AI 消息额度 | AI クレジット | | Usage Priority | 使用优先级 | 使用優先度 | | API Key | API Key | API キー | diff --git a/writing-guides/glossary.md b/writing-guides/glossary.md index 0e7982c73..b23fab367 100644 --- a/writing-guides/glossary.md +++ b/writing-guides/glossary.md @@ -34,6 +34,7 @@ Terms appear in body text exactly as written in this table. Capitalize them furt |:--------|:--------|:---------|:------| | model | 模型 | モデル | | | model provider | 模型供应商 | モデルプロバイダー | | +| AI credits | AI 消息额度 | AI クレジット | Dify Cloud credits for using supported models. Use "AI credits" in body prose. Apply normal title case in headings: "AI Credits" as a standalone term and singular "AI Credit" as a modifier, as in "AI Credit Usage." | | LLM (large language model) | 大语言模型 | 大規模言語モデル | | | chat model | 对话模型 | チャットモデル | Models that support role-based conversations (System/User/Assistant) | | completion model | 文本续写模型 | 補完モデル | Models designed for simple text continuation | @@ -43,8 +44,8 @@ Terms appear in body text exactly as written in this table. Capitalize them furt | rerank model | 重排序模型 | リランクモデル | Models that reorder retrieval results by relevance | | reasoning model | 推理模型 | 推論モデル | Models that output thinking process before final response | | moderation model | 内容审核模型 | モデレーションモデル | Models that detect and filter inappropriate content | -| TTS model | 文字转语音模型 | TTSモデル | Text-to-speech models | -| Speech2Text model | 语音转文字模型 | 音声認識モデル | Speech-to-text models | +| speech-to-text model | 语音转文字模型 | 音声認識モデル | Generic term; use "Speech-to-Text Model" only for the exact UI label. | +| text-to-speech model | 文字转语音模型 | TTSモデル | Generic term; use "Text-to-Speech Model" only for the exact UI label. | | token | token | token | | | API token | API 令牌 | APIトークン | | | prompt | 提示词 | プロンプト | | @@ -226,8 +227,8 @@ Terms in this section must match the Dify product interface exactly. When these | Rerank Model | Rerank 模型 | Rerank モデル | common.modelProvider.rerankModel.key | | | Speech-to-Text Model | 语音转文本模型 | 音声-to-テキストモデル | common.modelProvider.speechToTextModel.key | | | Text-to-Speech Model | 文本转语音模型 | テキスト-to-音声モデル | common.modelProvider.ttsModel.key | | -| Load Balancing | 负载均衡 | 負荷分散 | common.modelProvider.loadBalancing | | -| AI credits | AI 消息额度 | AI クレジット | common.modelProvider.card.aiCreditsOption | Renamed from "Message Credits" in v1.13.1. Lowercase c, as the UI renders it; the section header is the all-caps `modelProvider.quotaLabel`. | +| Load balancing | 负载均衡 | 負荷分散 | common.modelProvider.loadBalancing | | +| AI credits | AI 消息额度 | AI クレジット | common.modelProvider.card.aiCreditsOption | UI option matching the general term. Renamed from "Message Credits" in v1.13.1; the section header is the all-caps `modelProvider.quotaLabel`. | | Usage Priority | 使用优先级 | 使用優先度 | common.modelProvider.card.usagePriority | New in v1.13.1. Determines fallback order between API Key and AI credits. | | API Key | API Key | API キー | common.modelProvider.card.apiKeyOption | New in v1.13.1. Model provider credential option alongside AI credits. | @@ -479,4 +480,3 @@ Terms in this section must match the Dify product interface exactly. When these | Chunking Setting | 分段模式 | チャンキングモード | dataset-documents.embedding.mode | Section heading | | High-quality mode | 高质量模式 | 高品質モード | dataset-documents.embedding.highQuality | | | Economy mode | 经济模式 | 経済モード | dataset-documents.embedding.economy | | - From b760470cd4002290d4934dbbc89be256329ee9d5 Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 16:34:27 +0800 Subject: [PATCH 14/17] docs: clarify AI credit usage and API key terms --- en/cloud/use-dify/workspace/ai-credits.mdx | 24 ++++++----- .../use-dify/workspace/model-providers.mdx | 40 +++++++++---------- tools/translate/termbase_i18n.md | 1 + writing-guides/glossary.md | 1 + 4 files changed, 35 insertions(+), 31 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 0d252adc4..8737dd106 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -3,7 +3,7 @@ title: AI Credits description: Learn how AI credits work in your workspace, from model costs to usage priority --- -Each Dify Cloud plan includes a set number of AI credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API Keys. +Each Dify Cloud plan includes a set number of AI credits for using supported models hosted by Dify. They give you immediate access to models from different providers, so you can start building and testing apps without first adding your own API keys. App runs that use AI credits draw from the workspace's shared balance, including runs from the editor, published web apps, the Service API, and triggers. @@ -11,18 +11,20 @@ App runs that use AI credits draw from the workspace's shared balance, including Your workspace may use AI credits whenever a supported model runs. Common cases include: -- **Running an app with an LLM**, including models configured directly in Chatbot, Text Generator, and Agent apps or in LLM, Question Classifier, Parameter Extractor, and Agent nodes within Workflows and Chatflows. +- **Running an app with an LLM**, whether configured directly in a Chatbot, Text Generator, or Agent app or through an LLM, Question Classifier, Parameter Extractor, or Agent node in a Workflow or Chatflow. - **Using AI-assisted generation in the app editor** to create or revise prompts, code, structured output configurations, or Workflow and Chatflow drafts. -- **Indexing or searching a knowledge base**, which can involve an embedding model for indexing or vector search, an LLM when generating Q&A chunks from source documents, and a rerank model when reranking is enabled. +- **Indexing or searching a knowledge base**, which may run an embedding model for indexing or vector search, an LLM to generate Q&A chunks from source documents, or a rerank model when reranking is enabled. - **Transcribing voice input or generating audio output in an app** with a speech-to-text or text-to-speech model. -Each model available through AI credits has a fixed cost, regardless of input or output length. Dify deducts that cost each time it is used, rather than once per app run. +A single app run may involve several model runs when a Workflow has multiple model nodes, an Iteration or Loop repeats a node, or an Agent takes multiple reasoning steps. Knowledge base indexing may also run an embedding model in batches. -Suppose you index documents in a knowledge base, then test a Workflow that searches it with vector search. Indexing runs a 5-credit embedding model three times and uses 15 AI credits. The Workflow then runs the embedding model once and a 5-credit LLM twice, bringing the total to 30 AI credits. +Each model has a fixed credit cost, regardless of input or output length. That cost applies every time the model runs. + +For example, a Workflow uses an Iteration to process three items. A 5-credit LLM runs once per item, then once more to summarize the results. The four model runs use 20 AI credits in total. ## Compare Model Costs -The following costs apply each time a model runs: +Costs are grouped by provider: @@ -61,7 +63,7 @@ The following costs apply each time a model runs: -To use a model that is not listed above, [add your own API Key for its provider](/en/cloud/use-dify/workspace/model-providers#use-your-own-api-key). +Models not listed above are not available through AI credits. [Add your own API keys](/en/cloud/use-dify/workspace/model-providers#use-your-own-api-key) to use them. ## Check Your AI Credits @@ -82,11 +84,11 @@ After upgrading to Professional or Team, your new plan's AI credits are used bef ## Set Usage Priority -For a provider that supports AI credits, the first API Key you add is automatically used before AI credits. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: +For a provider that supports AI credits, the first API key you add is automatically used before AI credits. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: -- **AI credits first**: Requests use AI credits until they run out, then switch to the selected API Key. If the remaining balance cannot cover the model's full cost, change the priority to **API Key**. -- **API Key first**: Requests use the selected API Key. If it is unavailable when a request starts, the request uses available AI credits. +- **AI credits first**: Requests use AI credits until they run out, then switch to the selected API key. If the remaining balance cannot cover the model's full cost, change the priority to **API Key**. +- **API key first**: Requests use the selected API key. If it is unavailable when a request starts, the request uses available AI credits. -If a request has already been sent with your API Key and fails, it is not retried with AI credits. Fix the API Key or provider issue, or change the priority to **AI credits** before retrying. +If a request has already been sent with your API key and fails, it is not retried with AI credits. Fix the API key or provider issue, or change the priority to **AI credits** before retrying. diff --git a/en/cloud/use-dify/workspace/model-providers.mdx b/en/cloud/use-dify/workspace/model-providers.mdx index 40cc704b3..142b23353 100644 --- a/en/cloud/use-dify/workspace/model-providers.mdx +++ b/en/cloud/use-dify/workspace/model-providers.mdx @@ -1,33 +1,33 @@ --- title: Model Providers -description: Use models with your AI credits, or connect your own provider accounts and API Keys +description: Use models with your AI credits, or connect your own provider accounts and API keys --- -Models power many of the apps and features you build in Dify. Dify Cloud provides two ways to access them: use supported models with the AI credits included in your plan, or add your own API Key to connect a provider account. +Models power many of the apps and features you build in Dify. Dify Cloud provides two ways to access them: use supported models with the AI credits included in your plan, or add your own API key to connect a provider account. Both options let apps across the workspace access the provider's available models. ## Use AI Credits -With AI credits, you can use supported models without adding API Keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. +With AI credits, you can use supported models without adding API keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. ## Use Your Own API Key -Add your own API Key to access models through your provider account. +Add your own API key to access models through your provider account. -Only the workspace owner and admins can add or manage provider API Keys. +Only the workspace owner and admins can add or manage provider API keys. 1. Go to **Integrations** > **Model Provider**. 2. Find the provider. If it is not installed, select it under **Install model providers** or find it in the [Marketplace](https://marketplace.dify.ai/). -3. Click **Add API Key**, then enter your API Key and any other required details. +3. Click **Add API Key**, then enter your API key and any other required details. If the provider supports AI credits, both options remain available. Set [Usage Priority](/en/cloud/use-dify/workspace/ai-credits#set-usage-priority) to choose which one is used first. -Requests made with your API Key go through your provider account and do not deduct AI credits. For providers that do not support AI credits, an API Key is the only option. +Requests made with your API key go through your provider account and do not deduct AI credits. For providers that do not support AI credits, an API key is the only option. ## Add a Custom Model @@ -36,7 +36,7 @@ A provider's models are ready as soon as you connect it, so you only add one by Click **Add Model** on the provider's card, then give the model a name and its credentials. Providers that serve only a fixed set of models don't offer this option. -If the model you add matches the name and type of an existing one, Dify attaches the new API Key to that model instead of creating a duplicate. +If the model you add matches the name and type of an existing one, Dify attaches the new API key to that model instead of creating a duplicate. ## Choose the Models Apps Use by Default @@ -51,13 +51,13 @@ Apps and nodes that don't pick a model fall back to your workspace defaults. Cli ## Manage API Keys -Add more than one API Key to a provider when you need to switch between provider accounts or configure load balancing for a model. +Add more than one API key to a provider when you need to switch between provider accounts or configure load balancing for a model. -- For the models listed on a provider card, click **Configure** on the card to manage the API Keys they share. +- For the models listed on a provider card, click **Configure** on the card to manage the API keys they share. ![Manage Credentials for a Provider's Models](/images/use-dify/workspace/predefined-model-credential.png) -- For a custom model you added, click **Configure** on that model to manage its own API Keys. +- For a custom model you added, click **Configure** on that model to manage its own API keys. ![Custom Model Configuration](/images/use-dify/workspace/custom-model-config.png) @@ -68,16 +68,16 @@ Add more than one API Key to a provider when you need to switch between provider - A custom model relies on its own API Keys. Deleting its only API Key removes the model too. + A custom model relies on its own API keys. Deleting its only API key removes the model too. - To view every custom model's API Keys in one place, click **Manage Credentials** on the provider card. + To view every custom model's API keys in one place, click **Manage Credentials** on the provider card. ![Manage Credentials for All Custom Models](/images/use-dify/workspace/custom-model-credential-list.png) - API Keys stay here even after the custom model is removed, so you can re-add the model later without re-entering them. + API keys stay here even after the custom model is removed, so you can re-add the model later without re-entering them. ![Re-add a Removed Model](/images/use-dify/workspace/removed-custom-model-re-add.png) @@ -85,24 +85,24 @@ Add more than one API Key to a provider when you need to switch between provider ## Spread Requests Across API Keys with Load Balancing Professional Team -When one API Key handles many requests at once, it can hit the provider's rate limit and start failing. +When one API key handles many requests at once, it can hit the provider's rate limit and start failing. -Use load balancing to spread requests across several API Keys for the same model, so no single API Key becomes the bottleneck. Dify rotates through the API Keys in turn and rests any that hits its limit for a minute before trying it again. +Use load balancing to spread requests across several API keys for the same model, so no single API key becomes the bottleneck. Dify rotates through the API keys in turn and rests any that hits its limit for a minute before trying it again. 1. Find the model in the list, click **Configure**, and select **Load balancing**. -2. Click **Add credential** to add API Keys to the pool. +2. Click **Add credential** to add API keys to the pool. - Add a higher-quota or faster API Key more than once to send more of the traffic its way. + Add a higher-quota or faster API key more than once to send more of the traffic its way. ![Add Credentials for Load Balancing](/images/use-dify/workspace/add-load-balancing-credential.png) -3. Turn on at least two API Keys, then click **Save**. A model using load balancing shows a marker in the list. +3. Turn on at least two API keys, then click **Save**. A model using load balancing shows a marker in the list. -Switch back to a single API Key anytime; your load balancing setup is kept for later use. +Switch back to a single API key anytime; your load balancing setup is kept for later use. diff --git a/tools/translate/termbase_i18n.md b/tools/translate/termbase_i18n.md index bc86a3efb..82e60b2d0 100644 --- a/tools/translate/termbase_i18n.md +++ b/tools/translate/termbase_i18n.md @@ -29,6 +29,7 @@ | model | 模型 | モデル | | model provider | 模型供应商 | モデルプロバイダー | | AI credits | AI 消息额度 | AI クレジット | +| API key | API Key | API キー | | LLM (large language model) | 大语言模型 | 大規模言語モデル | | chat model | 对话模型 | チャットモデル | | completion model | 文本续写模型 | 補完モデル | diff --git a/writing-guides/glossary.md b/writing-guides/glossary.md index b23fab367..1e5d1df7e 100644 --- a/writing-guides/glossary.md +++ b/writing-guides/glossary.md @@ -35,6 +35,7 @@ Terms appear in body text exactly as written in this table. Capitalize them furt | model | 模型 | モデル | | | model provider | 模型供应商 | モデルプロバイダー | | | AI credits | AI 消息额度 | AI クレジット | Dify Cloud credits for using supported models. Use "AI credits" in body prose. Apply normal title case in headings: "AI Credits" as a standalone term and singular "AI Credit" as a modifier, as in "AI Credit Usage." | +| API key | API Key | API キー | Model provider credential. Use "API key" in body prose. Apply normal title case in headings: "API Key." The exact UI label is also "API Key." | | LLM (large language model) | 大语言模型 | 大規模言語モデル | | | chat model | 对话模型 | チャットモデル | Models that support role-based conversations (System/User/Assistant) | | completion model | 文本续写模型 | 補完モデル | Models designed for simple text continuation | From 7b0abfb6e6b166ff19a0f336621a83b377a83d28 Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 17:35:34 +0800 Subject: [PATCH 15/17] docs: refine AI credit usage guidance --- en/cloud/use-dify/workspace/ai-credits.mdx | 12 +++++------- en/cloud/use-dify/workspace/model-providers.mdx | 6 ++++-- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index 8737dd106..b4a07501f 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -16,15 +16,13 @@ Your workspace may use AI credits whenever a supported model runs. Common cases - **Indexing or searching a knowledge base**, which may run an embedding model for indexing or vector search, an LLM to generate Q&A chunks from source documents, or a rerank model when reranking is enabled. - **Transcribing voice input or generating audio output in an app** with a speech-to-text or text-to-speech model. -A single app run may involve several model runs when a Workflow has multiple model nodes, an Iteration or Loop repeats a node, or an Agent takes multiple reasoning steps. Knowledge base indexing may also run an embedding model in batches. +Each model has a fixed credit cost that applies every time it runs, regardless of input or output length. Iterations, Loops, Agent reasoning, and knowledge base indexing may run a model more than once. -Each model has a fixed credit cost, regardless of input or output length. That cost applies every time the model runs. - -For example, a Workflow uses an Iteration to process three items. A 5-credit LLM runs once per item, then once more to summarize the results. The four model runs use 20 AI credits in total. +For example, a Workflow uses a 5-credit LLM in an Iteration to analyze three customer reviews, then uses the same LLM to summarize the findings. The four runs use 20 AI credits in total. ## Compare Model Costs -Costs are grouped by provider: +The following costs apply each time a model runs: @@ -68,9 +66,9 @@ Models not listed above are not available through AI credits. [Add your own API ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to check how many AI credits your workspace has used out of the current total. If no credits remain, the page shows the date they ran out. On Professional and Team plans, you can also check the next renewal date. +Go to **Integrations** > **Model Provider** to check how many AI credits your workspace has used out of the current total. On Professional and Team plans, the page also shows the exact date of the next renewal. If no credits remain, it shows the date they ran out. -The number of included AI credits depends on your plan: +The table below shows the included credits and renewal schedule for each plan: | Plan | Included AI Credits | Renewal | |:-----|:--------------------|:--------| diff --git a/en/cloud/use-dify/workspace/model-providers.mdx b/en/cloud/use-dify/workspace/model-providers.mdx index 142b23353..510f67b94 100644 --- a/en/cloud/use-dify/workspace/model-providers.mdx +++ b/en/cloud/use-dify/workspace/model-providers.mdx @@ -5,11 +5,13 @@ description: Use models with your AI credits, or connect your own provider accou Models power many of the apps and features you build in Dify. Dify Cloud provides two ways to access them: use supported models with the AI credits included in your plan, or add your own API key to connect a provider account. -Both options let apps across the workspace access the provider's available models. +Models available through either option can be used across the workspace. ## Use AI Credits -With AI credits, you can use supported models without adding API keys. See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. +With AI credits, you can use supported models from different providers without adding API keys. Install one of these providers from **Integrations** > **Model Provider** (or the [Marketplace](https://marketplace.dify.ai/)), and its models are ready to use. + +See [AI Credits](/en/cloud/use-dify/workspace/ai-credits) for credit costs and usage rules. ## Use Your Own API Key From 9d03d9bf2c778cb397566e7cd52258436ab31c0c Mon Sep 17 00:00:00 2001 From: yajing Date: Wed, 19 Aug 2026 17:48:51 +0800 Subject: [PATCH 16/17] docs: refine AI credit status and priority wording --- en/cloud/use-dify/workspace/ai-credits.mdx | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index b4a07501f..c2c4877ab 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -66,9 +66,9 @@ Models not listed above are not available through AI credits. [Add your own API ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to check how many AI credits your workspace has used out of the current total. On Professional and Team plans, the page also shows the exact date of the next renewal. If no credits remain, it shows the date they ran out. +Go to **Integrations** > **Model Provider** to check how many AI credits your workspace has used out of the current total. If no credits remain, the page shows the date they ran out. On Professional and Team plans, you can also check the next renewal date. -The table below shows the included credits and renewal schedule for each plan: +The number of included AI credits depends on your plan: | Plan | Included AI Credits | Renewal | |:-----|:--------------------|:--------| @@ -85,8 +85,8 @@ After upgrading to Professional or Team, your new plan's AI credits are used bef For a provider that supports AI credits, the first API key you add is automatically used before AI credits. To change the order, go to **Integrations** > **Model Provider**, click **Configure** on the provider card, and set **Usage Priority**: - **AI credits first**: Requests use AI credits until they run out, then switch to the selected API key. If the remaining balance cannot cover the model's full cost, change the priority to **API Key**. -- **API key first**: Requests use the selected API key. If it is unavailable when a request starts, the request uses available AI credits. +- **API key first**: Requests use the selected API key. If a request cannot be sent with that key, it uses available AI credits. -If a request has already been sent with your API key and fails, it is not retried with AI credits. Fix the API key or provider issue, or change the priority to **AI credits** before retrying. +If a request has already been sent with your API key but fails, it is not retried with AI credits. Fix the API key or provider issue, or change the priority to **AI credits** before retrying. From 3507ebc3981b373ec007d8f72b3291a2240d0957 Mon Sep 17 00:00:00 2001 From: yajing Date: Mon, 24 Aug 2026 11:09:05 +0800 Subject: [PATCH 17/17] docs: add xAI credit costs and usage guidance --- en/cloud/use-dify/workspace/ai-credits.mdx | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/en/cloud/use-dify/workspace/ai-credits.mdx b/en/cloud/use-dify/workspace/ai-credits.mdx index c2c4877ab..c2440d08d 100644 --- a/en/cloud/use-dify/workspace/ai-credits.mdx +++ b/en/cloud/use-dify/workspace/ai-credits.mdx @@ -49,6 +49,12 @@ The following costs apply each time a model runs: - **3 credits**: `Gemini 3 Flash-Preview`. - **1 credit**: `Gemini 3.5 Flash-Lite`, `Gemini 3.1 Flash-Lite-Preview`, `Gemini 3.1 Flash-Lite`, `Gemini 2.5 Flash`, `Gemini 2.0 Flash`, `Gemini 2.0 Flash-Lite`. + + + + - **20 credits**: `grok-4.5`, `grok-4.6`. + - **10 credits**: `grok-4.3`. + @@ -66,7 +72,7 @@ Models not listed above are not available through AI credits. [Add your own API ## Check Your AI Credits -Go to **Integrations** > **Model Provider** to check how many AI credits your workspace has used out of the current total. If no credits remain, the page shows the date they ran out. On Professional and Team plans, you can also check the next renewal date. +Check your remaining AI credits below your workspace name. Click the balance to open **Integrations** > **Model Provider** for usage details. The number of included AI credits depends on your plan: