product updateOpenAI

OpenAI shifts Codex to usage-based pricing, offers $500 credits to enterprise customers

TL;DR

OpenAI is replacing per-seat licensing with usage-based pricing for Codex in ChatGPT Business and Enterprise plans, eliminating upfront license costs. Eligible Business customers can claim up to $500 in promotional credit per workspace. The shift targets enterprises where coding tools typically expand from individual developers to full teams, positioning OpenAI against GitHub Copilot and Cursor.

1 min read
0

OpenAI Shifts Codex to Usage-Based Pricing for Enterprise Plans

OpenAI is moving Codex to a usage-based pricing model across ChatGPT Business and Enterprise plans, removing per-seat licensing requirements and requiring payment only for actual usage.

The change applies to both ChatGPT Business and Enterprise tiers. Workspace administrators can enable Codex access across their organization and pay exclusively for consumption—no upfront license commitment required. As part of a limited-time promotion, eligible Business customers can claim up to $500 in promotional credit per workspace.

Strategic Rationale

OpenAI frames the shift as lowering adoption barriers for enterprises. "Coding tools typically spread from individual developers to full teams," the company stated. "This model gives organizations a simpler way to support that motion inside a managed workspace."

The pricing change directly targets GitHub Copilot and Cursor, both of which continue charging per-seat subscription fees. By eliminating upfront costs, OpenAI aims to reduce friction for teams evaluating coding assistants at scale.

Usage Context

OpenAI reports that over two million developers use Codex weekly. Business and Enterprise plan usage has grown sixfold since January 2026, according to the company.

The primary competitive threat comes from Anthropic's Claude Code offering, which OpenAI identified as its biggest rival in this category.

What This Means

OpenAI is applying a proven SaaS playbook: lower initial friction to drive adoption, then monetize based on actual consumption. For enterprises, this removes a key barrier—the inability to enable coding tools company-wide without pre-purchasing licenses. The $500 credit effectively subsidizes initial trial usage.

The move assumes lock-in increases with usage; developers embedded in a coding assistant tend to stick with it. However, success depends on Codex's quality relative to Copilot and Cursor. Pricing advantage alone won't retain customers if product quality lags.

Related Articles

benchmark

GPT-6.2 Astra and Sol cut answer length 34 to 40 percent but fall sharply on Admits Uncertainty

An Artificial Analysis evaluation finds OpenAI's GPT-6.2 Astra and Sol produce answers 34 to 40 percent shorter than their predecessors and lead with the answer more often. Both score lower on uncertainty and transparency, and trail Anthropic's Claude Fable 5.2 and Opus 5.5 on the Transparency and Organization indexes.

changelog

OpenAI tells developers to rewrite system prompts for GPT-6.2 Astra and Sol as defaults shift

OpenAI says developers moving to GPT-6.2 Astra or Sol should review their system prompts. The company published copy-ready templates in the models' documentation, and the Responses API now discards earlier reasoning content if conversation history is edited.

model release

OpenAI releases GPT-Live-1.1 in the API, with standby mode billing idle WebRTC time at $0.02 per minute

OpenAI has released GPT-Live-1.1 in the API. The full-duplex voice model improves multilingual accents and transcription, and a new standby mode bills idle WebRTC time at $0.02 per minute instead of $0.05.

benchmark

GPT-6.2 Astra scores 41.2% on XDOF's Ember-0 robot benchmark, up from 19.6% for GPT-6 Astra

Robotics infrastructure company XDOF reports that GPT-6.2 Astra reached a 41.2% average success rate across the 16 tasks of its Ember-0 benchmark, versus 19.6% for GPT-6 Astra. It is the first model to complete rope knot-tying and T-shirt unfolding on the benchmark. Average prediction latency fell from 15 seconds to 10.1 seconds.

Comments

Loading...