DeepSeek hasn't been SotA in at least 12 calendar months, which might as well be a decade in LLM years

What about Kimi and GLM?

These are well behind the general state of the art (1yr or so), though they're arguably the best openly-available models.

	▲	tgrowazay 30 minutes ago \| parent [-]
		According to artificial analysis ranking, GLM-5 is at #4 after Claude Opus 4.5, GPT-5.2-xhigh and Claude Opus 4.6 .