> ## Documentation Index
> Fetch the complete documentation index at: https://www.traceloop.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Sessions

> Group traces from multi-turn conversations into sessions

<Frame>
  <img className="block dark:hidden" src="https://mintcdn.com/enrolla/8JEbCsSWBgIb14pj/img/conversation/conversation-light.png?fit=max&auto=format&n=8JEbCsSWBgIb14pj&q=85&s=c9ffffbb6d3a3fe98531955b81009f62" width="2438" height="1229" data-path="img/conversation/conversation-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/enrolla/8JEbCsSWBgIb14pj/img/conversation/conversation-dark.png?fit=max&auto=format&n=8JEbCsSWBgIb14pj&q=85&s=ea168b9f3e4b6c80b0fcd590e583bcea" width="2438" height="1229" data-path="img/conversation/conversation-dark.png" />
</Frame>

Sessions let you group related traces from a multi-turn conversation into a single view.
When you set a conversation ID on your traces, the Traceloop UI groups them into a session where you can see the full conversation flow and inspect individual spans.

<Tip>
  Use a consistent conversation ID across all turns of the same conversation.
  Generate a unique ID (e.g. a UUID) when the conversation starts and pass it on each subsequent interaction.
</Tip>

## Setting a Conversation ID

<Tabs>
  <Tab title="Python">
    ### Using the decorator

    ```python theme={null}
    from traceloop.sdk.decorators import conversation, workflow

    @workflow(name="chat_interaction")
    @conversation(conversation_id="session-123")
    async def handle_message(user_message: str):
        response = client.chat.completions.create(
            model="gpt-4o",
            messages=[{"role": "user", "content": user_message}],
        )
        return response.choices[0].message.content
    ```

    ### Using the direct API

    You can also set the conversation ID directly, without a decorator:

    ```python theme={null}
    from traceloop.sdk.tracing import set_conversation_id

    set_conversation_id("session-123")
    ```
  </Tab>

  <Tab title="Typescript">
    ### Using the decorator

    ```typescript theme={null}
    import * as traceloop from "@traceloop/node-server-sdk";

    class Chatbot {
      @traceloop.conversation("session-123")
      async handleMessage(userMessage: string) {
        const response = await openai.chat.completions.create({
          model: "gpt-4o",
          messages: [{ role: "user", content: userMessage }],
        });
        return response.choices[0].message.content;
      }
    }
    ```

    ### Using `withConversation`

    ```typescript theme={null}
    import * as traceloop from "@traceloop/node-server-sdk";

    await traceloop.withConversation("session-123", async () => {
      const response = await openai.chat.completions.create({
        model: "gpt-4",
        messages: [{ role: "user", content: userMessage }],
      });
      return response.choices[0].message.content;
    });
    ```

    ### Using `conversationId` in workflow config

    You can also pass `conversationId` as part of the workflow or task configuration:

    ```typescript theme={null}
    import * as traceloop from "@traceloop/node-server-sdk";

    await traceloop.withWorkflow(
      { name: "chat_interaction", conversationId: "session-123" },
      async () => {
        const response = await openai.chat.completions.create({
          model: "gpt-4",
          messages: [{ role: "user", content: userMessage }],
        });
        return response.choices[0].message.content;
      }
    );
    ```
  </Tab>
</Tabs>