Quite hidden in the mobile app at least. I had to click thinking. The slider goes 5.6 Instant, 5.6 Medium, 5.6 High, 5.6 Extra High, and the surprise 6 Pro
Shouldn't this new reduction in thinking output make the model cheaper to operate? Could it be that the model is the same old LLM, with a bit newer architecture but still doing the same things, including huge amounts of thinking, just that its thinking is less clear to human readers?
Quite hidden in the mobile app at least. I had to click thinking. The slider goes 5.6 Instant, 5.6 Medium, 5.6 High, 5.6 Extra High, and the surprise 6 Pro
I see this as well. Odd there is only 6 Pro.
Also is rolling out to the $100/$200 Codex subscriptions, I got access this morning.
Unsurprisingly, it does indeed consume usage at ~2.5x the rate of Sol.
Shouldn't this new reduction in thinking output make the model cheaper to operate? Could it be that the model is the same old LLM, with a bit newer architecture but still doing the same things, including huge amounts of thinking, just that its thinking is less clear to human readers?
Thinking output is already a relatively small part of the total input compared to raw code/text files.