baseline deep profile

Codex

OpenAI's coding agent for completing software-engineering work across ChatGPT, apps, CLI, IDE and cloud workflows.

ProviderOpenAI

Catalog profileAvailable

Completeness38%

LifecycleActive

Checked2026-09-02

Core record

Status: Public

Access: ChatGPT / App / CLI / IDE / Cloud

API status: Product and developer integrations

Openness: Closed

Modalities: Code and text

Best suited for

End-to-end engineering tasks, refactors, migrations, issue work and parallel coding-agent workflows.

Limitations

Availability, included usage and supported environments vary by ChatGPT plan and product surface.

API / auth

Interface: Pending field audit

Base URL: Pending

Authentication: Pending field audit

Compatibility: Pending field audit

Rate limits: Pending field audit

Deployment

ChatGPT / App / CLI / IDE / Cloud

Models / versions

Model	Context	Max output	Modalities	License	Status
Pending dedicated model/version audit.

Pricing snapshots

SKU / model	Unit	Input	Output	Flat	Notes
Pending dedicated pricing audit. Artificial Index does not infer current prices.

Integrations

Pending

Regions

Pending regional availability audit.

Governance / privacy

Pending governance/privacy audit.

Benchmarks

Pending methodology-aware benchmark audit unless separately documented. Vendor and independent benchmark results are kept distinct.

Field verification matrix

identity: verifiedorganization: verifiedcategories: verifiedstatus: partialaccess: partialapi: partialauthentication: pendingmodels: pendingcontext: pendingmax_output: pendingmodalities: partialopenness_license: partialpricing: pendingrate_limits: pendingdeployment: partialregions: pendingintegrations: pendinggovernance: pendingbenchmarks: pendinglifecycle: verifiedsources: verified

Evidence

Codex primary public sourceFields: identity, organization, categories, status, access, modalities, openness_license, lifecycle, sources · confidence 90% · checked 2026-09-02V1 baseline source anchor. Catalog presence is source-linked; unresolved technical fields still require dedicated review.Source ↗

V1 rule: unknown technical details remain pending. Catalog presence is not proof of current pricing, API limits, model lineage, regional availability, privacy terms or benchmark performance.