PromptFu vs promptfoo — What Is the Difference?

If you searched for “promptfoo” and landed here, you might be wondering: is this the official promptfoo site? Are PromptFu and promptfoo the same thing? Here’s the clear answer.

The Short Version

promptfoo PromptFu
What it is Open-source CLI framework Independent knowledge hub
Domain promptfoo.dev (acquired by OpenAI) promptfu.com
Purpose Run automated LLM prompt tests Tutorials, guides, community resources
Status Now part of OpenAI (March 2026) Independent, vendor-neutral
Affiliation OpenAI (post-acquisition) Community-run, no vendor affiliation

What Is promptfoo (the framework)?

Promptfoo is an open-source framework for testing, evaluating, and red-teaming AI prompts. It started as a developer tool to solve a real problem: as LLM applications move to production, you need a way to verify that your prompts produce correct, safe, and consistent outputs.

The name is a programming pun — “foo” is the classic placeholder variable (foo, bar, baz) applied to prompt engineering. A prompt is your input to an AI model; foo signals it’s the first, most important thing to test.

With promptfoo, you write test cases in YAML, define expected outputs or graders, and run:

promptfoo eval

It generates a report showing which prompts passed, which failed, and how outputs changed across model versions. Think of it as pytest or Jest — but for LLM prompts.

Where does promptfoo live now?

The promptfoo project (promptfoo.dev) was acquired by OpenAI in March 2026 for approximately $40 million. The core team joined OpenAI, and the framework is being integrated into OpenAI’s developer tooling.

The open-source repository continues to exist, but active community development has shifted under OpenAI’s umbrella.


What Is PromptFu (this site)?

PromptFu (this site, at promptfu.com) is an independent publication covering:

  • Prompt engineering techniques and best practices
  • LLM evaluation strategies across providers
  • AI developer tools including promptfoo tutorials
  • Cheat sheets and reference guides
  • Framework comparisons (promptfoo, Braintrust, RAGAS, etc.)

We are not affiliated with OpenAI or the original promptfoo team. Our goal is to be a vendor-neutral resource that remains useful regardless of how the commercial tools evolve.


Why Does PromptFu Exist?

When the promptfoo.dev acquisition was announced, many in the community wanted an independent destination — a place not controlled by OpenAI or any single company — where prompt engineering knowledge and best practices would live on.

PromptFu fills that role:

  • promptfoo.dev → The official OpenAI-owned tool and documentation
  • promptfu.com → The PromptFu knowledge hub (this site) — independent, vendor-neutral

Should I Use promptfoo After the OpenAI Acquisition?

Yes — promptfoo remains an excellent tool. The acquisition doesn’t change the core capability: define test cases, run evaluations, get a report. If you’re building on OpenAI’s models, the integration will likely get even tighter.

What changes is the independence question. If you want prompt evaluation that works equally well with Anthropic, Google, Mistral, and open-source models — and you want documentation from a source with no vendor bias — that’s where PromptFu comes in.


The PromptFu Alternative Perspective

PromptFu covers the full prompt engineering landscape, not just one tool. Our take on prompt evaluation includes:


Summary

  • promptfoo (lowercase) = the testing framework, now owned by OpenAI
  • PromptFu = this site, an independent resource for the prompt engineering community
  • promptfu.com = where PromptFu lives

Both are valuable depending on what you need: the tool or the knowledge.

Ready for another?
Best AI Developer Tools in 2026

The AI developer tooling landscape has matured significantly. There are fewer “revolutionary” announcements and more tools that have earned their place in daily workflows through sustained reliability and genuine productivity gains.

This is not a comprehensive list of every AI tool—it is a curated set of the ones that are worth your time in 2026, organized by category.

AI prompts, command-line cheat sheets & developer tips — prompt engineering guides, LLM evaluation tools, and AI tools for developers building with modern language models.

 2026