View and customize trace display - Weights & Biases Documentation

After you create Calls in W&B Weave, you often want to open a single call to inspect its inputs, outputs, and metadata. This page shows how to view a call in the UI or in the SDK, and how to customize how Weave renders trace data in the UI using weave.Markdown. Use the following tabs to choose the interface that fits your workflow.

Web App
Python
TypeScript
HTTP API

To view a Call in the UI:

Navigate to wandb.ai and select your project.
In the Weave project sidebar, click Traces.
Find the Call you want to view in the table.
Click the Call to open its details page.

For details on the Trace view, see Navigate the Weave Trace view.

To view a call using the W&B Weave Python SDK, use the get_call method:

import weave

# Initialize the client
client = weave.init("your-project-name")

# Get a specific call by its ID
call = client.get_call("call-uuid-here")

print(call)

import * as weave from 'weave'

// Initialize the client
const client = await weave.init('intro-example')

// Get a specific call by its ID
const call = await client.getCall('call-uuid-here')

console.log(call)

To view a call using the Service API, make a request to the /call/read endpoint.

curl -L 'https://trace.wandb.ai/call/read' \
-H 'Content-Type: application/json' \
-H 'Accept: application/json' \
-d '{
    "project_id": "string",
    "id": "string",
}'

Customize rendered traces with `weave.Markdown`

After you can view a Call, you may want to control how its inputs and outputs appear in the UI. Use weave.Markdown to customize how Weave displays your trace information without losing the original data. This lets you render your inputs and outputs as readable blocks of formatted content while preserving the underlying data structure.

Python
Typescript

Use postprocess_inputs and postprocess_output functions in your @weave.op decorator to format your trace data. The following code sample uses postprocessors to render a call in Weave with more readable formatting:

import weave

def postprocess_inputs(query) -> weave.Markdown:
    search_box = f"""
**Search Query:**
``+`
{query}
``+`
"""
    return {"search_box": weave.Markdown(search_box),
            "query": query}

def postprocess_output(docs) -> weave.Markdown:
    formatted_docs = f"""
# {docs[0]["title"]}

{docs[0]["content"]}

[Read more]({docs[0]["url"]})

---

# {docs[1]["title"]}

{docs[1]["content"]}

[Read more]({docs[1]["url"]})
"""
    return weave.Markdown(formatted_docs)

@weave.op(
    postprocess_inputs=postprocess_inputs,
    postprocess_output=postprocess_output,
)
def rag_step(query):
    # example newspaper articles of the companies on the S&P 500 
    docs = [
        {
            "title": "OpenAI",
            "content": "OpenAI is a company that makes AI models.",
            "url": "https://www.openai.com",
        },
        {
            "title": "Google",
            "content": "Google is a company that makes search engines.",
            "url": "https://www.google.com",
        },
    ]
    return docs

if __name__ == "__main__":
    weave.init('markdown_renderers')
    rag_step("Tell me about OpenAI")

This feature is not available in the TypeScript SDK yet.

After you run the op, Weave renders the formatted Markdown alongside the original data in the Call’s details page. The following screenshot shows the unformatted and formatted outputs side by side.

​Customize rendered traces with weave.Markdown

Customize rendered traces with `weave.Markdown`