OpenAI ships GPT-6 Sol and Luna at half prior API cost; GitHub adds Copilot access

The mid-tier GPT-6 models join Astra with cheaper API pricing and staged ChatGPT and Copilot rollouts; Anthropic released an Opus update the same day, TechCrunch said.

Confirmed By Priya Nair · Staff Writer — Technology & Science United States / Global Published Sep 23, 2026, 1:41 PM PT Updated Sep 23, 2026, 1:45 PM PT

OpenAI on Tuesday expanded its GPT-6 lineup with updated Sol and Luna models, positioning them as cheaper, more efficient companions to GPT-6 Astra, which the company released earlier in September, TechCrunch (opens in new tab) reported.

OpenAI said Sol targets complex work such as coding, while Luna is aimed at high-volume tasks like summarizing documents, extracting information, or answering quick questions. TechCrunch (opens in new tab) reported that the GPT-6 Sol and Luna API prices are half those of the 5.6-series predecessors, a cut OpenAI attributed to caching and inference improvements. The company also claimed GPT-6 Sol makes about half as many mistakes as its predecessor on an internal factuality evaluation based on de-identified user-flagged conversations—a company metric, not an independent benchmark.

Availability is staged. TechCrunch (opens in new tab) said the models are in ChatGPT Work and Codex for most paid accounts and in the ChatGPT API, with Luna reaching Free and Go users in the desktop app and a gradual ChatGPT app and website rollout. Separately, GitHub’s changelog said GPT-6 Sol and Luna are available in GitHub Copilot under usage-based billing, with Sol for Pro+, Max, Business, and Enterprise plans and Luna also for Copilot Pro, across Visual Studio Code, Visual Studio, Copilot CLI, JetBrains, Xcode, Eclipse, and other surfaces, with a gradual rollout.

TechCrunch (opens in new tab) noted that Anthropic released a new version of Opus 5.5 about 90 minutes before OpenAI’s announcement, framing the timing as competitive pressure. OpenAI’s comparative claims that Sol and Luna outperform Anthropic models such as Fable and Opus are the company’s marketing assertions and are not independently verified here.