OpenAI released two new artificial intelligence models, GPT-6 Sol and GPT-6 Luna, on Tuesday, cutting API prices in half compared with GPT-5.6's promotional rates. The launch came just minutes after rival Anthropic introduced Claude Opus 5.5, its own new flagship model. The back-to-back releases underscored the escalating competition between the two leading AI companies—and handed developers and businesses that pay by the token an unusually direct head-to-head on price and performance.

Sol and Luna sit below GPT-6 Astra, the model OpenAI called "the most intelligent and aligned model in the world" when it launched on September 3. Astra remains OpenAI's top choice for the hardest jobs; Sol and Luna are the cheaper, faster options built for everyday work.

The tiering strategy closely mirrors the lineup Anthropic has applied: Astra is the equivalent of Fable, Sol corresponds to Opus, Terra parallels Sonnet, and Luna matches Haiku once it receives its own upgrade. The parallel structure also makes tier-for-tier price comparison straightforward.

Internal safety metrics also improved, according to the company: the coding-deception rate fell to 1.3% for Sol and 2.8% for Luna, both far below GPT-5.6 Sol's 10.4%.

OpenAI framed the price cuts as a way to remain competitive. API costs for both models were reduced by 50% versus GPT-5.6's promotional per-token rates. Tokens are the chunks of text a model reads and writes—usually slightly less than a full word—and companies pay by the million, which is how AI bills add up at scale.

OpenAI announced the launch in a post on X:

Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe. GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster and more affordable models to support work at scale. We've also made caching and inference more efficient, and… pic.twitter.com/5LiVE4rbFt

— OpenAI (@OpenAI) September 22, 2026

The post was published on OpenAI's official X account (@OpenAI) on September 22, 2026.

Sol now costs $2 per million input tokens and $10 per million output tokens, down from $4 and $20. Luna drops to $0.10 and $0.50, down from $0.20 and $1.20. Overall, OpenAI is the cheaper offer per tier when compared against Anthropic.

The performance case leans on AutomationBench, a benchmark built by Zapier that tests whether an AI agent can carry out a full business workflow using 47 tools spanning sales, marketing, operations, support, finance, and HR, scored as a pass rate. GPT-6 Sol at its highest reasoning setting scored 33.2% for $0.27 per task—where halved token rates translate directly into per-task savings, since agent workloads consume tokens across every step of a job.

Claude Opus 5, at its own top setting, scored 26.9% but cost more than 11 times as much per task, according to OpenAI's own numbers. GPT-6 Sol also edged out Claude Fable 5.1, Anthropic's pricier flagship, on the same test. Not, the figures reference Claude Opus 5 rather than the Opus 5.5 Anthropic released minutes earlier, so the newest Claude model does not yet appear in the comparisons.

On Agents' Last Exam, which evaluates AI agents on long, economically valuable work across 55 sub-industries and grades results as a percentage completed, GPT-6 Sol at its highest effort scored 56.4%. OpenAI says that beats Claude Opus 5's best score at a 60% lower cost per task.

Reasoning effort is a dial OpenAI added: turn it up, and the model spends more time and computing power double-checking itself before answering, which tends to raise accuracy and cost together.

For computer use—an AI agent clicking, typing, and navigating software the way a person would—GPT-6 Sol on OSWorld 2.0 nearly tied Claude Opus 5's medium-effort score, 60.5% to 60.3%, at about 80% less cost. OSWorld tests long, realistic computer tasks and reports a partial score based on how much of the job was completed correctly. These head-to-head figures are drawn from OpenAI's own reporting, and independent benchmark results have not yet been published.

GPT-6 Sol and Luna are live now in ChatGPT Work and Codex for Plus, Pro, Business, Enterprise, and Edu subscribers, with Luna also reaching free and Go users through the desktop app. Neither model is in the plain ChatGPT app yet, and OpenAI says it is rolling both out gradually throughout the day to keep the service stable. From here, the visible open slots in an otherwise mirrored rivalry are the models' arrival in the plain ChatGPT app and the Haiku-tier counterpart still awaiting its own upgrade on Anthropic's side.