Pipeline & ask()

ask() and run_pipeline() POST the native pipeline JSON. Flags for cache, RAG, compression, PII, and context layers match the console pipeline.

python

out = ts.ask(

    "Summarize in 3 bullets.",

    provider="openai",

    model="gpt-4o",

    use_cache=True,

print(out.text, out.metrics.cost_usd)

For field-level parity with the UI, see the full method reference (Pipeline options card).