> 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/advanced-basic/chat/model-compare-branches.md).

# Multi-Model Comparison and Message Branches

Multi-model comparison is suitable for problems without a single correct answer, such as solution reviews, copy direction, and cross-checking materials. Message branching lets you try another route from a certain node without duplicating the entire conversation.

## Compare multiple models at the same time

{% stepper %}
{% step %}

### 1. Open \[Chat] and click the model name

Select the models you want to compare in the model picker. Before using it for the first time, first make sure the services these models belong to can connect normally.
{% endstep %}

{% step %}

### 2. Send the same question

Include the evaluation criteria in the question, for example, “Compare from the perspectives of feasibility, risk, and cost.” Don’t just ask, “Which is better?”
{% endstep %}

{% step %}

### 3. Compare the differences, rather than just choosing the longest answer

Focus on whether the facts are consistent, whether the assumptions are clear, what has been omitted, and which option better fits your constraints. Important facts should still be checked against the original source material.
{% endstep %}
{% endstepper %}

<figure><img src="/files/6e480690b987061aa8346a25a82a2b1637d647d2" alt="对话中的模型选择器和多个模型入口"><figcaption><p>Select multiple models from the model picker, then use the same question to compare the differences.</p></figcaption></figure>

{% hint style="warning" %}
Selecting multiple models at once will send separate requests. When fees, speed, or sensitive data are involved, first use a short question to confirm the connection and results, then handle longer materials.
{% endhint %}

## Create a branch from a message

Find the message you want to explore again, open the message menu, and choose the branch option. The new branch will retain the previous context, and subsequent messages will be recorded separately from the original path. Use the branch manager to switch between paths, compare them, and go back.

You can also create an empty branch in the branch canvas. After creation, an empty branch is saved immediately, remains after restarting the app, and stays on the branch canvas; the next time you send something in the input box, that branch will be filled in. Empty branches you no longer need can be deleted from the node’s right-click menu.

<figure><img src="/files/18cec66a47a2ff00434d8096a1228b8f85ab0e97" alt="分支管理器中两条对话分支和六个消息节点"><figcaption><p>The branch manager keeps both the pre-launch checklist and the quick pilot routes.</p></figcaption></figure>

In the figure: ① branch nodes and the current path; ② legends for users, assistants, current path, and deactivated paths. The example retains the two routes “Pre-launch Checklist” and “Quick Pilot,” with a total of 2 branches and 6 message nodes.

### Use case: reviewing two release plans

First ask the model to identify the risks and gaps in the plans, then follow up from the same reply with “add the pre-launch checklist” and “evaluate from the quick pilot perspective.” After opening the branch manager, the two routes will be kept side by side, and you can continue asking questions on each separately or return to the other route at any time to verify conclusions.

<details>

<summary>When is it not suitable to open multiple models?</summary>

If you are only looking up one clear fact, organizing a short text, or the material contains content that should not be sent to multiple service providers, using a single model is more appropriate.

</details>

<details>

<summary>Will branching modify the original message?</summary>

No. A branch continues from the selected node, and the original path remains intact and can be returned to at any time.

</details>


---

# 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/advanced-basic/chat/model-compare-branches.md?ask=<question>&goal=<endgoal>
```

`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.