Yes, honestly, Opus 4.6 and GPT 5.4 were mostly not really noticeable improvements over 4.5 and 5.3 respectively. If we were stuck at 4.5 levels but at 1/10th of the price, I'll take it.

▲

furyofantares 4 hours ago | parent [-]

I find 4.6 pretty noticeable upgrade, but it might be the 1M context. I'm interested in how the 1M context works out with Qwen.

	▲	Alifatisk 2 hours ago \| parent \| next [-]
		From Qwen-3-max thinking, I remember the inference becoming veeery slow as you pushed towards 1M context, already at 300k tokens you would notice the degradation. But of course, I was using Qwen Chat, so could be a resource allocation thing.
	▲	nwienert 4 hours ago \| parent \| prev [-]
		I found it worse, in a very clear way.