Load Vast.ai Whisper ASR data to DuckDB

Vast.ai is a GPU cloud marketplace providing a REST API for managing compute resources and specialized services like Whisper ASR deployments. Everything needed to build a working Vast.ai Whisper ASR → DuckDB pipeline is on this page: the API's base URL, authentication, endpoints, pagination and incremental field — plus a prompt that hands the whole job to your coding agent.

---

Build your Vast.ai Whisper ASR to DuckDB pipeline

Paste this prompt into Claude, Codex, or Cursor. The agent does the rest.

PromptCopy prompt
Run uvx dlthub-init@latest to build a pipeline from Vast.ai Whisper ASR to DuckDB and run it on dltHub

That scaffolds a dltHub workspace and installs the dltHub AI harness — the project rules, the secrets-management skill, and the dlt MCP server your agent needs to work safely. From there it reads the Vast.ai Whisper ASR API, proposes the endpoints to load, then writes, runs and validates the pipeline while you review rather than type. Credentials are inspected through MCP tools, so your agent never reads secrets.toml itself. How the LLM-native workflow works →

Prefer to write it yourself? Every fact the agent uses is below.

---

Vast.ai Whisper ASR API at a glance

These values come from the Vast.ai Whisper ASR API reference — the authoritative source if anything here looks out of date.

---

How do I authenticate with the Vast.ai Whisper ASR API?

All API requests require the Authorization header with the format 'Authorization: Bearer <API_KEY>'.

1. Get your credentials

To obtain your API credentials for Vast.ai, follow these steps: 1. Navigate to the Vast.ai web console and log in. 2. Go to the 'Keys' section (or 'Manage Keys' page). 3. Click the '+New' button to initiate the creation of a new API key. 4. Assign a descriptive name to the key and configure any necessary permission scopes if required. 5. Click create/save and copy the generated API key immediately, as it will only be displayed once. Treat this key as a secure secret.

2. Add them to .dlt/secrets.toml

Copy
[sources.vast_ai_whisper_asr_source]
vast_api_key = "your_actual_api_key_here"

dlt reads this file automatically at runtime. With the harness, the setup-secrets skill prompts you for the values and never handles the raw credential in chat. For production, see setting up credentials with dlt.

---

What Vast.ai Whisper ASR data can I load into DuckDB?

These are the Vast.ai Whisper ASR endpoints dlt can load into DuckDB:

---

How do I load only new Vast.ai Whisper ASR records?

Vast.ai Whisper ASR exposes after_token on api/v0/instances/, so dlt can request only the records that changed since the last run. Set it as the cursor_path and dlt tracks the high-water mark for you between runs.

Copy
{"name": "instances", "endpoint": {
    "path": "api/v0/instances/",
    "data_selector": "instances",
    "incremental": {"cursor_path": "after_token", "initial_value": "2024-01-01T00:00:00Z"},
}}

On the first run dlt loads everything from initial_value; on every run after that it requests only what changed and appends with write_disposition="merge" if you set a primary key. See incremental loading.

---

What does the generated Vast.ai Whisper ASR pipeline look like?

A standard dlt REST API pipeline — the same code you would write by hand, loading /detect-language and /asr (alternatively, some deployments utilize /transcription) from the Vast.ai Whisper ASR API into DuckDB:

Copy
import dlt
from dlt.sources.rest_api import RESTAPIConfig, rest_api_resources

@dlt.source
def vast_ai_whisper_asr_source(api_key=dlt.secrets.value):
    config: RESTAPIConfig = {
        "client": {
            "base_url": "https://console.vast.ai/api/v0",
            "auth": {"type": "bearer", "token": api_key},
        },
        "resources": [
            {"name": "instances", "endpoint": {"path": "api/v0/instances/", "data_selector": "instances"}},
            {"name": "charges", "endpoint": {"path": "api/v0/billing/charges/", "data_selector": "charges"}}
        ],
    }
    yield from rest_api_resources(config)


def load_vast_ai_whisper_asr_to_duckdb() -> None:
    pipeline = dlt.pipeline(
        pipeline_name="vast_ai_whisper_asr_pipeline",
        destination="duckdb",
        dataset_name="vast_ai_whisper_asr_data",
    )
    load_info = pipeline.run(vast_ai_whisper_asr_source())
    print(load_info)


if __name__ == "__main__":
    load_vast_ai_whisper_asr_to_duckdb()

Run it with python vast_ai_whisper_asr_pipeline.py. The agent iterates on this until it loads cleanly — you review and approve, rather than write it from scratch.

---

How do I query Vast.ai Whisper ASR data in DuckDB?

dlt creates one table per resource. Query the loaded data with Python or SQL — or ask your agent to, through the MCP server's execute_sql_query tool.

Python (pandas DataFrame):

Copy
import dlt

data = dlt.pipeline("vast_ai_whisper_asr_pipeline").dataset()
df = data.instances.df()
print(df.head())

SQL:

Copy
SELECT * FROM vast_ai_whisper_asr_data.instances LIMIT 10;

See querying your data with dataset and exploring it in marimo notebooks.

---

How do I deploy the Vast.ai Whisper ASR to DuckDB pipeline in production?

The pipeline runs locally, which is ideal for prototyping and one-off analysis. When you need it on a schedule, monitored on every load, and shared with your team, deploy the same dlt code on the dltHub platform — no infrastructure to maintain. The prompt above already ends with "run it on dltHub", so your agent can take it there directly.

• Deploy & schedule — run the pipeline as a managed job with automatic retries.
• Monitor — observable job queues, alerting, and load metrics for every run.
• Transform — promote raw Vast.ai Whisper ASR loads into governed, documented models.
• Visualize & share — explore data in notebooks and publish live dashboards instead of static screenshots.

Book a demo →

---

What other destinations can I load Vast.ai Whisper ASR data to?

dlt loads into any of these — only the destination argument changes:

Set dlt.pipeline(destination="snowflake") and add credentials in .dlt/secrets.toml. On the dltHub platform the same pipeline runs against a managed Iceberg lakehouse. See the full destinations list.

---

Next steps

• Ready for production? Book a demo of the dltHub platform →
• dlt LLM-native workflow docs
• REST API source reference
• Join the dlt community on Slack