> For the complete documentation index, see [llms.txt](https://docs.cherryai.com.cn/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.cherryai.com.cn/docs/en-us/cherry-studio/installation/v1-v2-feature-differences.md).
# Feature differences
V2 adjusted the data structure and multiple feature entry points. Before upgrading, first see the items that need reconfiguration; see details in [\[Upgrade from V1 to V2\]](/docs/en-us/cherry-studio/installation/v1-to-v2-migration.md).
## V2.0.2 migration changes
| Situation | How V2.0.2 is handled | Notes |
| ----------------------- | ------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
| First migration from V1 | You can install V2.0.2 directly from V1.9.13 | There is no need to install V2.0.0 first. |
| Normal V2 upgrade | Upgrade directly and continue using the current V2 data | Do not click \[Re-migrate]. |
| Re-migrate V1 | Select \[Re-migrate] under \[Settings] → \[Data] | The current V2 data will be permanently deleted; use this only for migration failures or missed data from before. |
{% hint style="danger" %}
\[Re-migrate] will not add and merge V1 data into the current V2. It will first delete the current V2 data, then re-import from the retained V1 data. Do not click it unless the V1 migration failed or data was missed.
{% endhint %}
## How the data is handled
| processing method | Data scope | What to do after upgrading |
| --------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------- |
| Automatic migration | Settings, model services and models, assistants and groups, sessions and messages, Agents and sessions, MCP, knowledge base and valid indexes, files, drawing, translation, notes, prompts, usage records | Spot-check whether commonly used items can be opened normally. |
| Needs review | Anthropic credentials, default service for web search, Agent tool permissions and knowledge base bindings, prompt order, knowledge base failed items, custom CSS, sidebar favorites, missing files | Reconfirm according to the table below. |
| Will not be inherited | Agent scheduled task history, legacy sitemap expansion results, some temporary image references, current open state of notes, configuration for removed features | Recreate when needed. |
{% hint style="info" %}
Migration will not delete the original V1 data. New V2 data will not sync back to V1, and backups of the two versions cannot be restored to each other.
{% endhint %}
## Main differences
| Function | V1 | V2 | What to do after upgrading |
| ------------------------------ | ----------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- |
| Assistants and prompts | Independent assistant library; quick phrases can be associated with assistants | Assistants are managed in chat and Agent; quick phrases are merged into global prompts | Check assistant groups and prompt order. |
| Agent | Some configurations and workspaces follow the session; old authorizations can automatically approve tools | Identity, memory, and main configuration are handled by the Agent; the workspace is fixed after the session is created; tools may request authorization again | Check the model, workspace, tools, and permissions. |
| Agent knowledge base | Knowledge tools may access the global knowledge base | Only explicitly bound knowledge bases are accessed | Recheck knowledge base bindings for each Agent. |
| Knowledge base retrieval | You can manually select the retrieval mode and sitemap source | BM25 is used when there is no embedding model; hybrid retrieval is used when there is an embedding model; sitemaps are treated as normal URLs | Check the embedding and reranking models; rebuild failed sources. |
| Web search | You can select the service in the assistant or input area; includes local search, RAG compression, and subscription blacklist | Configure keyword search and URL reading separately under \[Settings] → \[Web Search]; related old options have been removed | Re-select the two default services and check the credentials. |
| MCP | Added services and third-party discovery markets coexist | Added services continue to migrate; the old third-party discovery market is no longer available | Check service status; new services are added from the current marketplace or JSON. |
| Model services | Anthropic can save OAuth credentials | Anthropic OAuth is not migrated; AWS Bedrock regions are retained; newly added services are off by default | Re-enter the API key for Anthropic and enable services as needed. |
| Files | The relationship between file copies and business objects is weak | Hosted files are managed by reference; hosted copies are cleaned up with a delay after the last reference is deleted | Files to be kept long term are placed in \[Files] or exported; user original files will not be deleted. |
| Drawing, Mini App, and sidebar | Parameters and entry points are in the legacy area | Drawing parameters have moved to the prompt toolbar; Mini App is in the top tab bar; sidebar favorites are reset | Familiarize yourself with the new entry points and reset your favorites. |
| Custom CSS | V1 selector takes effect directly | The content is retained, but disabled by default after migration | Enable it again after adapting it to the V2 selector. |
| Code CLI | iFlow can be selected | iFlow is replaced by Qoder | Use Qoder instead when related workflows are needed. |
## Developer compatibility
What changes are there in the API and external integrations?
* The model identifier of API Gateway changed from `provider::model` to `provider:model`.
* Knowledge API returned entries use V2 fields.
* The MCP-over-HTTP endpoint has been removed and does not affect in-app MCP.
* If SSE startup fails, a normal HTTP error may be returned directly; clients need to support both HTTP and SSE error paths.
* The scheduled task interface parameters have changed, and historical run records will not be migrated.
## References
* [Upgrade V1 to V2](/docs/en-us/cherry-studio/installation/v1-to-v2-migration.md)
* [Cherry Studio V2 official download](https://cherryai.com.cn/download)
* V2.0.2 release page:[GitCode](https://gitcode.com/CherryHQ/cherry-studio/releases/v2.0.2) · [GitHub](https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.2)
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.cherryai.com.cn/docs/en-us/cherry-studio/installation/v1-v2-feature-differences.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.