Pricing / Dify Cloud
Built to Grow with What You Build.
Sandbox
For trying out Dify's core features.
Start for FreeFree
- 200 message creditsMessage credits are provided to help you easily try out different models from OpenAI, Anthropic, Gemini, xAI, DeepSeek, Tongyi in Dify. Credits are consumed based on the model type. Once they're used up, you can switch to your own API key.
- 1 Team Workspace
- 1 Team Member
- 5 Apps
- 50 Knowledge DocumentsQuota on the number of documents imported from the Knowledge Data Source.
- 50MB Knowledge Data StorageDocuments with the High Quality indexing mode will consume Knowledge Data Storage resources. When Knowledge Data Storage reaches the limit, new documents will not be uploaded.
- 10 Knowledge Request/minSpecifies the total number of actions a workspace can perform per minute within the knowledge base, including dataset creation, deletion, updates, document uploads, modifications, archiving, and knowledge base queries. This metric is used to evaluate the performance of knowledge base requests. For example, if a Sandbox user performs 10 consecutive hit tests within one minute, their workspace will be temporarily restricted from performing the following actions for the next minute: dataset creation, deletion, updates, and document uploads or modifications.
- Standard Document Processing
- 3,000 Trigger EventsThe number of events that automatically start workflows through Plugin, Schedule, or Webhook triggers.
- Up to 2 Triggers/workflow
- Standard Workflow ExecutionWorkflow execution queue priority and speed.
- 10 Annotation Quota LimitsManual editing and annotation of responses provides customizable high-quality question-answering abilities for apps. (Applicable only in Chat apps)
- 30 Days Log History
- 5,000 API Rate Limit/month
- 200 message credits
Professional
PopularFor independent developers and small teams building production AI apps.
Start Building$590per workspace / year- 5,000 message credits / monthMessage credits are provided to help you easily try out different models from OpenAI, Anthropic, Gemini, xAI, DeepSeek, Tongyi in Dify. Credits are consumed based on the model type. Once they're used up, you can switch to your own API key.
- 1 Team Workspace
- 3 Team Members
- 50 Apps
- 500 Knowledge DocumentsQuota on the number of documents imported from the Knowledge Data Source.
- 5GB Knowledge Data StorageDocuments with the High Quality indexing mode will consume Knowledge Data Storage resources. When Knowledge Data Storage reaches the limit, new documents will not be uploaded.
- 100 Knowledge Request Rate Limit/minSpecifies the total number of actions a workspace can perform per minute within the knowledge base, including dataset creation, deletion, updates, document uploads, modifications, archiving, and knowledge base queries. This metric is used to evaluate the performance of knowledge base requests. For example, if a Sandbox user performs 10 consecutive hit tests within one minute, their workspace will be temporarily restricted from performing the following actions for the next minute: dataset creation, deletion, updates, and document uploads or modifications.
- Priority Document Processing
- 20,000 Trigger Events / monthThe number of events that automatically start workflows through Plugin, Schedule, or Webhook triggers.
- Unlimited Triggers / workflow
- Faster Workflow ExecutionWorkflow execution queue priority and speed.
- 2,000 Annotation Quota LimitsManual editing and annotation of responses provides customizable high-quality question-answering abilities for apps. (Applicable only in Chat apps)
- Unlimited Log History
- No Dify API Rate Limit
- 5,000 message credits / month
Team
For medium-sized teams needing collaboration and higher throughput.
Get Started$1590per workspace / year- 10,000 message credits / monthMessage credits are provided to help you easily try out different models from OpenAI, Anthropic, Gemini, xAI, DeepSeek, Tongyi in Dify. Credits are consumed based on the model type. Once they're used up, you can switch to your own API key.
- 1 Team Workspace
- 50 Team Members
- 200 Apps
- 1,000 Knowledge DocumentsQuota on the number of documents imported from the Knowledge Data Source.
- 20GB Knowledge Data StorageDocuments with the High Quality indexing mode will consume Knowledge Data Storage resources. When Knowledge Data Storage reaches the limit, new documents will not be uploaded.
- 1,000 Knowledge Request Rate Limit/minSpecifies the total number of actions a workspace can perform per minute within the knowledge base, including dataset creation, deletion, updates, document uploads, modifications, archiving, and knowledge base queries. This metric is used to evaluate the performance of knowledge base requests. For example, if a Sandbox user performs 10 consecutive hit tests within one minute, their workspace will be temporarily restricted from performing the following actions for the next minute: dataset creation, deletion, updates, and document uploads or modifications.
- Top Priority Document Processing
- Unlimited Trigger EventsThe number of events that automatically start workflows through Plugin, Schedule, or Webhook triggers.
- Unlimited Triggers / workflow
- Priority Workflow ExecutionWorkflow execution queue priority and speed.
- 5,000 Annotation Quota LimitsManual editing and annotation of responses provides customizable high-quality question-answering abilities for apps. (Applicable only in Chat apps)
- Unlimited Log History
- No Dify API Rate Limit
- 10,000 message credits / month
Compare Plans & Features
Overview of what's included in our different plans.
SandboxFree | Professional$59Per workspace/monthSave $118 with annual billing | Team$159Per workspace/monthSave $318 with annual billing | |
|---|---|---|---|
| Model Provider | |||
| LLM API Load Balancing Adds multiple API keys to models, effectively bypassing the API rate limits. | |||
| Model Vendor Management Supports adding and configuring API keys from different model providers for the workspace. | |||
| App Studio | |||
| Chatbot | |||
| Text Generator | |||
| Agent | |||
| Chatflow | |||
| Workflow | |||
| Webapp Icon Customization | |||
| App Version Control Allows users to manually save versions and restore to any version. | Coming Soon | Coming Soon | |
| Explore Page Templates Provides beginners with high-quality, scenario-based templates to quickly learn and build applications with Dify. | |||
| Dify Marketplace | Free for Now | Free for Now | Free for Now |
| Knowledge Base (RAG) | |||
| Batch Document Upload Allows bulk document upload to the knowledge base. | Up to 50 Files/Websites per Upload | Up to 50 Files/Websites per Upload | |
| Knowledge Base Sources / Customize Knowledge Supports various data sources: import text, use external sources (Notion or sync Web content), and connect external knowledge bases. | Standard / External Knowledge API | Standard / External Knowledge API | Standard / External Knowledge API |
| Pipeline Templates Customize and share reusable knowledge pipeline templates to standardize processes, reduce setup time, and improve team efficiency. | |||
| Add Chunks | |||
| App Deploy | |||
| Publish as Webapp Supports publishing AI applications built in Dify as web apps for public access on the internet. | |||
| Publish as API | |||
| App Monitor & Feedback | |||
| App Runtime Data Analysis | Standard, Advanced(Langsmith/ Langfuse..) | Standard, Advanced(Langsmith/ Langfuse..) | Standard, Advanced(Langsmith/ Langfuse..) |
| App Logs History | 30 Days | Unlimited | Unlimited |
| Workspace Management | |||
| WebApp Branding Customization Supports custom LOGO and UI. | |||
| Role Management Supports four roles (Owner, Admin, Editor, Normal). Only the Owner and Admin can manage roles, billing, and models. Additionally, they can view operation logs of each individual workspace member. | |||
| Support | |||
| Support | Community, Help docs | Priority Email | Priority Email |
| SOC Type II Report | |||
We've Got Answers.
Here's everything you need to know before getting started.
Can I try Dify without a paid subscription?
You can try Dify for free under the Sandbox plan. Within a limited number of users, you can experience all features of the current version until resource limits are reached. When vector space, number of apps, etc. reach the limit, the system will remind you to upgrade to a higher plan to continue using more resources.
How do message credits work?
Message credits can be used with the following models, each consuming different credits per AI response:
OpenAI
- gpt-5.6-sol / gpt-5.5: 20 credits
- gpt-5.6-terra / gpt-5.4 / gpt-5.2 / gpt-4o: 10 credits
- gpt-5.6-luna / gpt-5.4-mini / gpt-5 / gpt-5.1 / gpt-4.1 / o3-mini: 5 credits
- gpt-5-mini / gpt-4.1-mini / gpt-4o-mini / gpt-3.5-turbo: 1 credit
Anthropic
- Claude Opus 4.7 / Claude Opus 4.6 / Claude Opus 4.5: 20 credits
- Claude Sonnet 4.6 / Claude Sonnet 4.5: 10 credits
- Claude Haiku 4.5: 5 credits
Gemini
- Gemini 3.5 Flash / Gemini 3.1 Pro-Preview: 10 credits
- Gemini 3 Flash-Preview: 3 credits
- Gemini 3.1 Flash-Lite-Preview / Gemini 3.1 Flash-Lite / Gemini 2.5 Flash / Gemini 2.0 Flash / Gemini 2.0 Flash-Lite: 1 credit
xAI
- grok-4.3: 5 credits
Tongyi
- 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: 10 credits
- qwen3.5-plus / qwen3.6-flash / qwen-flash / qwq-plus / qwen3.5-omni-flash / qwen3-vl-plus / qwen-vl-max: 5 credits
- qwen3.7-plus / qwen3.5-flash / qwen-turbo / qwen-long-latest / qwen-long-2025-01-25 / qwen3-vl-flash / qwen-vl-plus: 1 credit
You can switch models in settings. Once message credits are exhausted, you can continue with your own LLM API key.
*Note: Due to model costs, some models cannot be used with message credits. Please use your own API key to access them.
Who is responsible for team payments?
Only the team owner and admins can access the Billing Page and manage future payments for the team.
Can I pay for Dify Cloud Service subscription with a corporate bank account or other business payment method instead of a personal credit card for internal enterprise use?
We’re exploring Pay by Invoice via bank transfer, but it is not yet available. Under the proposed arrangement, it would be available only for annual Team subscriptions. Invoices would be issued by LangGenius, Inc.; payments from outside the United States would require an international wire transfer, with all related fees paid by the customer. Please contact us to share whether this option would meet your needs.
How do I cancel my paid plan?
You can cancel your plan by going to “Settings” → “Billing” and then clicking “Manage Billing and Subscriptions”. Cancellation will take effect at the end of the current billing cycle.
Where can I find my invoices and receipts?
Team owners and admins can access, view, and download your workspace’s invoices and receipts by going to “Settings” → “Billing” and clicking “Manage Billing and Subscriptions”.
If my Professional plan resources reach the limit, can I purchase more vector space or team seats independently?
We recommend upgrading your plan to meet your usage needs. If the Team plan doesn’t suffice, please contact support@dify.ai for help.
Is Dify secure? Where is my data stored?
Dify Cloud runs on hardened infrastructure with encryption in transit and at rest; data is stored in our managed cloud region.
How to obtain Self-hosted edition?
The open-source Community edition is free to self-host; reach out for Enterprise self-hosting options.
Still unsure? Email us at cloud@dify.ai for more information.
Ready to Build the AI App of Tomorrow?
Launch production-ready agents powered by RAG pipelines, integrations, and full observability - no heavy lifting required.