Compatible (BYOK)

Use Promptfoo with GlideflowAI.

An LLM evaluation tool that accepts a custom OpenAI API base URL. Connect it through GlideflowAI's OpenAI-compatible API with a key you control.

Get a key

What this connection is for

Promptfoo is an evaluation and testing tool for comparing prompts, model outputs, assertions, and application behavior. It is for developers who want a repeatable test suite around an LLM feature rather than deciding on a model from a few ad hoc chats. Its YAML configuration makes it practical to run the same cases locally and in CI when a prompt, model, or provider changes.

Choose Promptfoo when you need regression checks, red-team cases, or side-by-side provider evaluation. It is not an end-user chat interface or an orchestration framework. Keep evaluation keys in the CI secret store and make the provider identifier explicit, because ambiguous model routing can make a comparison less useful than it appears.

Promptfoo accepts OPENAI_API_BASE_URL, OPENAI_BASE_URL, and per-provider apiBaseUrl for OpenAI-compatible chat providers. Point those settings at GlideflowAI, use openai:chat with an exact model ID, and run a small fixture before a larger evaluation suite. Pin the provider ID in version control so CI uses the same evaluation target as local runs. Promptfoo is an independent project; this is a user-managed connection, not an endorsement or partnership.

Why pair it with GlideflowAI

Promptfoo can send the same evaluation cases to a specific GlideflowAI model by making the base URL and chat provider explicit in YAML.

That helps preserve a reproducible test baseline when the team changes prompts or compares catalog models.

Setup

Use a new API key for the tool, select a model from the catalog, and begin with a small task. Do not put a key in a repository, screenshot, or shared configuration file.

1
Set the compatible endpoint and key
Export both variables in the shell or CI secret environment that will execute evaluations.
```
export OPENAI_API_BASE_URL="https://api.glideflowai.com/v1"
export OPENAI_API_KEY="sk-your-key"
```

Declare an explicit chat provider

Use an explicit OpenAI Chat provider ID with an exact gateway model ID in promptfooconfig.yaml.

providers:
  - id: openai:chat:glm-5.2
    config:
      apiBaseUrl: https://api.glideflowai.com/v1
      apiKeyEnvar: OPENAI_API_KEY

Need the endpoint basics? Read authentication, browse the model catalog, or compare pricing.

Frequently asked questions

Which Promptfoo environment variable sets the full endpoint?

Use OPENAI_API_BASE_URL=https://api.glideflowai.com/v1; Promptfoo also documents OPENAI_BASE_URL, but the full API base variable is clearer for this setup.

Why use openai:chat:glm-5.2 instead of openai:glm-5.2?

The explicit chat prefix makes routing deterministic for a compatible Chat Completions endpoint rather than relying on Promptfoo's model-name inference.

Can I put apiKey directly in promptfooconfig.yaml?

Promptfoo supports it, but apiKeyEnvar is safer for shared configuration and CI because it keeps the secret outside the YAML file.

Is this an official integration?

No. The connection is configured by you using a compatible API and does not imply sponsorship, endorsement, or a partnership.

Ready to connect Promptfoo?

Create a key, use the setup above, and verify the connection on a small task before you scale it up.