If OpenAI has a new model that they are close to releasing, now seems like a perfect opening to steal some thunder. Mythos coming out later with only marginal improvements to a new OpenAI model would be good-great outcome for OpenAI
replyGemini and Codex already scored higher on benchmarks than Opus 4.6 and they recently added a $100 tier with limited 2x limits, that's their answer and it seems people have caught on.
reply> that's their answer and it seems people have caught on.
replyThere's nothing to catch on to. OpenAI have been shouting "come to us!! We are 10x cheaper than Anthropic, you can use any harness" and people don't come in droves. Because the product is noticeably worse.