Skip to content

Layer A · LLM Observability & Evaluation

Promptfoo

Part of OpenAI

Acquired

Open-source test harness that compares prompts and models on output quality, estimated cost and latency, with red teaming for security.

Acquired by OpenAI.

Last checked . Visit Promptfoo

At a glance

Control layer
A: Build and run
Status
Acquired by OpenAI
Typical buyers
Engineering and platform, AI leadership (CAIO)
Delivery
Open source
  • Evaluation
  • Red teaming

Spend controls it documents

Not a spend control

Buyer fit

Useful before a model or prompt change ships: run the same test cases across candidates and see what quality each one buys at what cost. Cost assertions fail a test when a response is too expensive; they do not cap spending in production. Best owned by engineering, with results shared with whoever approves model choices.

Market outlook

OpenAI announced on 9 March 2026 that it will acquire Promptfoo and fold it into its Frontier agent platform, while keeping the open-source project and support for other providers. Watch whether multi-provider testing stays first-class.

What to check

  • Cost and latency assertions apply to completed responses. They are a test gate, not a spend control.
  • Configuration is plain files in the repository, so switching cost is low.
  • Check how the pending OpenAI ownership affects support for competing model providers.

Ownership and name changes

  1. 9 March 2026

    OpenAI agrees to acquire Promptfoo

    The open-source evaluation and red-teaming tool is to be folded into OpenAI Frontier; OpenAI says the open-source project will continue. OpenAI announcement

Sources

The map describes vendors from public sources; it does not rate, rank or recommend them, and no listing is paid for. Confirm pricing and capabilities with the vendor. See how the map is built.

By Ed Epstein, who works at IBM. The postings on this site are my own and don't necessarily represent IBM's positions, strategies or opinions. About and disclosure · Methodology