✳ the wire · analysis
GPT-6.1 Sol ships at DevDay at $2/$10, and OpenAI says it nearly matches Astra
Get launch alerts like this, free →

OpenAI launched GPT-6.1 Sol at its DevDay keynote on September 29. It is available now in ChatGPT Work and Codex for Plus, Pro, Business, Enterprise and Edu users, not yet in Chat, and in the API as gpt-6.1-sol.
THE PRICE
$2 per million input tokens and $10 per million output, the same sticker as GPT-6 Sol. Cached input is $0.10 per million, 95% below standard input and half of GPT-6 Sol's cached rate. OpenAI says it nearly matches GPT-6 Astra on agentic coding, computer use and professional work at a fifth of Astra's standard token prices. That is the vendor's claim, from its own evaluations; we will report third-party numbers when they land.
WHAT IS COMING
GPT-6.1 Sol Ultrafast, with up to 8x faster token generation in Codex, "in the coming days", per OpenAI.
HOW WE CAUGHT IT, AND WHY IT WAS LATE
Our sweep read gpt-6.1-sol on OpenAI's public model docs at 17:23:45 UTC and the alert went out at 17:24:45, to 318 inboxes and one phone. That was about 20 minutes after the keynote announced it. The reason is ours: our OpenAI source reads OpenAI's own model list, which needs a key we do not have configured, and neither OpenRouter nor models.dev had listed it yet. We added OpenAI's public model docs as a source during the keynote, and that is what caught it. It stays on for every OpenAI launch from here.
The next Sol is armed on the cadence.
Source: OpenAI launch page (openai.com/index/introducing-gpt-6-1-sol), read directly · OpenAI's public model docs · our own sweep ↗ · GPT-6.1 Sol tracker · the bench index
