Breaking: OpenAI just retired a model that's only seven days old.
GPT-6 Sol launched, and before most people even finished testing it, GPT-6.1 Sol showed up to replace it.
One week of shelf life. That's the new normal apparently.
Here's what makes this wild: GPT-6.1 Sol isn't just newer, it's measurably smarter. It gained 4 points on the Intelligence Index over its predecessor.
It now sits just 1 point below GPT-6 Astra, OpenAI's flagship top-tier model, while costing less than a quarter of the Cost per Task.
Think of it like getting Ferrari-level performance out of a car that sips fuel at a quarter of the rate.
Pricing stayed the same on paper: $2 per million input tokens, $10 per million output tokens. But there's a subtle upgrade buried in there.
The cache read discount jumped from 90% to 95%. For agentic workloads (think AI agents running repetitive, chained tasks non-stop), that small shift actually lowers the real blended cost.
And this comes right after a 50% price cut from GPT-5.6 Sol to GPT-6 Sol just weeks earlier. This isn't just a smarter model, it's an aggressive pricing war with no brakes.
The AI race now moves faster than app update cycles. A week is apparently long enough for an entire model generation to become history.
🔗 Original Source & Reference: https://artificialanalysis.ai/articles/gpt-6-1-sol-replaces-gpt-6-sol-after-just-7-days-with-near-astra-intelligence
Published automatically via FeedMind AI Content Pipeline.
Top comments (0)