What is 'live steering'?

pi is deliberately small and lets you redirect it mid-run rather than waiting for a finished attempt. Whether that beats a fuller harness depends on how much of the work you want to direct yourself.

Does the smaller harness do less?

It does less around the model on purpose. That is an advantage when you want control and a disadvantage when you want the harness to plan; running the same task through both is the fastest way to tell which you are.

Can I change the options?

Yes. Change the available models or add another compatible agent in the composer before sending. All options use normal usage billing unless explicitly marked as sponsored.

How does blind comparison work?

You know the candidate set, but results are labelled anonymously until you submit your preference. The original sessions remain available after reveal.

This comparison uses normal compute and model pricing. Specific sponsored activities have their own limited coverage; comparing does not automatically make usage free.

Codex vs pi

Codex is OpenAI's full coding harness; pi is deliberately small and lets you steer it mid-run. Whether the extra machinery earns its keep depends on how much of the work you want to direct yourself.

RepositoryChoose a repository

How they have gone here

Published pairing: Codex vs pi. Human votes on this exact pairing, from blind matches readers judged — the evidence on this page describes that pairing, not whatever the composer above is currently set to.

The board counts a comparison only when it runs two different agents on the lineup's own models, so this combination has no record of its own. The published numbers below are measured independently.

See the full standings

Published reference numbers

What an independent lab measured for the models behind these options.

Intelligence v4.3

General capability across the lab's whole suite.

47.1

CodexGPT-5.6 Sol (max)

piGPT-5.6 Sol (max)

Coding

The coding subset of the same suite.

77.4

CodexGPT-5.6 Sol (max)

piGPT-5.6 Sol (max)

Agentic

Tasks the model runs in several steps.

50.5

CodexGPT-5.6 Sol (max)

piGPT-5.6 Sol (max)

These describe the model on its own, not the harness, the tools or the whole setup this page compares — and the lab runs each model at maximum effort, which is not what the composer above is set to. Scores are only comparable within one index version.

Source: Artificial Analysis — independent measurements, not ours, taken September 16, 2026.

View chart values

Measure	CodexGPT-5.6 Sol (max)	piGPT-5.6 Sol (max)
Intelligence index	47.1	47.1
Input$ / 1M tokens	4.00	4.00
Output$ / 1M tokens	20.00	20.00
Output speedtokens / s	64.7	64.7
Time to first tokens	94.96	94.96
Cost per task$	1.99	1.99

Source: Artificial Analysis — independent measurements, not ours, taken September 16, 2026. A dash is a measure they do not publish for that model.

Scored as a whole configuration

An independent benchmark that runs the harness and the model together, which is the unit this page compares.

Accuracy · Terminal-Bench 4.0

Percent of trials the whole configuration solved.

37.3%

Not published

CodexCodex · GPT-5.6 Sol123 of 330 trials · max effort

pipi · GPT-5.6 Sol

The whisker is the 95% confidence interval. Two configurations whose whiskers overlap are not separated by this benchmark, however far apart their bars look.

Source: Terminal-Bench — an independent harness-and-model benchmark, not ours, taken September 17, 2026.

View chart values

Configuration	AccuracyTerminal-Bench 4.0
CodexCodex · GPT-5.6 Sol	37.3% ± 3.8123 of 330 trials · max effort
pipi · GPT-5.6 Sol	No published run

± is the 95% confidence interval. Two configurations whose intervals overlap are not separated by this benchmark, however far apart their percentages look.

Source: Terminal-Bench — an independent harness-and-model benchmark, not ours, taken September 17, 2026. A dash is something they do not publish for that option.

Catalog price and context

What the open model catalog lists for the models behind these options.

Catalog	Codexopenai/gpt-5.6-sol	piopenai/gpt-5.6-sol
Input$ / 1M tokens	2.00	2.00
Output$ / 1M tokens	10.00	10.00
Context windowtokens	1.1M	1.1M
Providers serving it	3	3

List price for the model alone. What a run from this page costs also depends on the harness, the tools it calls and how long it works.

Source: OpenRouter — the open model catalog, not ours, taken September 17, 2026. A dash is something they do not publish for that option.

Your task, your choice

Compare the work you actually need.

Both options receive your task and attachments. The selected models and tools define this comparison; results on one task do not establish a universal winner.

Research
Compare two products using their official documentation. Ask for a recommendation, source links, and unresolved questions.
Writing
Give both the same brief and audience. Compare accuracy, clarity and the changes you would need before using the draft.
Build something
Describe a page or small application, then inspect the actual output and ask each agent to improve it.

Common questions

Related resources

All comparisons

Start with editable options.

Keep working with your preferred result.

Compare your task

Codex vs pi

How they have gone here

Published reference numbers

Scored as a whole configuration

Catalog price and context

Compare the work you actually need.

Research

Writing

Build something

What is 'live steering'?

Does the smaller harness do less?

Can I change the options?

How does blind comparison work?

Is this free?

All comparisons

Keep working with your preferred result.