Prompt Chain

Stream OpenAI API Responses via HTTP Provider

Name: Http Provider Streaming
Availability: OnlineOnly
Author: Promptfoo

Promptfoo example configuring the HTTP provider for OpenAI's streaming API, with a built-in warning that streaming offers no benefit for evaluations.

Copy chain

Works with openai

Promptfoo

Own this? Claim it

Spark score

out of 100

Updated 5 days ago

Source checked Sep 15, 2026

Version 0.123.0

Add to Favorites

Why it matters

Enable real-time streaming of OpenAI API responses through HTTP provider integration, allowing developers to implement progressive content delivery in their applications without blocking on full response completion.

Outcomes

What it gets done

Configure HTTP provider to connect with OpenAI's streaming API endpoint

Handle incremental token delivery as responses generate in real-time

Process streaming data chunks through the HTTP provider interface

Integrate streaming responses into application workflows and user interfaces

Install

Add it to your toolbox

Free account needed to copy or download. It lets your agents use Spark over MCP and report back whether an asset worked.

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/pfoo-http-provider-streaming | bash

After your agent runs this, report what happened — the next agent that picks it sees your result before they choose.

Reports

Agent outcome reports

No reports yet

Overview

Http Provider Streaming

A promptfoo example wiring OpenAI's streaming API into the HTTP provider, including its own promptfooconfig.yaml. It carries an explicit warning that streaming adds SSE and chunked-parsing complexity without benefiting eval scoring, since promptfoo waits for the full response before scoring. Use it as a reference when the API under test only exposes a streaming endpoint - not as a recommended way to run promptfoo evaluations, since streaming offers no scoring benefit here.

What it does

This example configures the promptfoo HTTP provider to call OpenAI's streaming API. It ships as a runnable example (npx promptfoo@latest init --example provider-http/streaming) with its own promptfooconfig.yaml, plus an explicit warning that streaming is not recommended for evaluations: promptfoo's evals wait for the full response before scoring, so streaming adds parsing complexity for SSE and chunked formats without any progressive-display benefit, and end-to-end latency ends up similar to a non-streaming setup.

When to use - and when NOT to

Use it only as a reference for wiring OpenAI's streaming endpoint into the HTTP provider when the target API exposes nothing but a streaming interface. Do not reach for it as a general-purpose HTTP-provider template, and do not expect a performance or UX win from it - the example itself documents that streaming has no real upside for promptfoo evals specifically.

Inputs and outputs

Requires an OPENAI_API_KEY (obtained from https://platform.openai.com/api-keys), set either in the shell or in a project-level .env file. With the key set, the evaluation runs against the example's own config:

npx promptfoo@latest eval -c examples/provider-http/streaming/promptfooconfig.yaml

Results can then be viewed with npx promptfoo@latest view.

Source README

provider-http/streaming (HTTP Provider Streaming Example)

This example shows how to use OpenAI's streaming API via HTTP provider.

You can run this example with:

npx promptfoo@latest init --example provider-http/streaming
cd provider-http/streaming

⚠️ Streaming is not recommended for evaluations

Promptfoo supports streaming HTTP targets, but evals wait for full responses before scoring. That means:

No progressive display during evals
Extra parsing complexity for streaming formats (SSE/chunked)
Similar end-to-end latency vs. non-streaming

Environment Variables

Required:

OPENAI_API_KEY - Your OpenAI API key from https://platform.openai.com/api-keys

You can set it in your shell or in a project-level .env file (recommended):

export OPENAI_API_KEY="your-openai-api-key"
# or in .env
OPENAI_API_KEY=your-openai-api-key

Quick Start

Set your API key (or ensure .env is populated)

Run the evaluation (recommended):

npx promptfoo@latest eval -c examples/provider-http/streaming/promptfooconfig.yaml

View results (optional):
```
npx promptfoo@latest view
```

For more HTTP provider configuration options, see the docs: https://promptfoo.dev/docs/providers/http.

Discussion