I’m surprised no one is commenting on how cheap this is compared to Opus 4.x and GPT-5.5.
$1.25 / $2.50 for every M input and output tokens.
Is this is a smaller less powerful model? What am I missing?
Grok is associated with Elon Musk. If we used $TSLA profit margin as a proxy, it looks like it's no longer as high. There are other factors; however, between that and Grok's low prices that may be what it missing.
Yes, it’s a significantly less powerful model, that’s why.
They dropped the output cost, butthe input cost is relatively high. This is a recent trend. Seen with DeepSeek 4 Pro as well.