Is Min P sampling really the preferred modern alternative to Top K/Top P?

According to what I've been reading (and also according to all models I've asked about this), the consensus seems to be that Min P is the better/more modern approach to sampling and that it should be preferred over Top P/Top K, which should be used only if Min P isn't available or for legacy reasons... Yet, looking and recently published LLM on huggingface and elsewhere, the recommended parameters for sampling are still largely Top K and/or Top P. Is this only for legacy reasons? Or some other r

More from Reddit

How to run a local coding agent with Gemma 4 and Pi | Patrick Loeber

Reddit

Anthropic's Claude remote uses Nested-If-Else-Statements-v8.4 [SATIRE]

Reddit

I’ve been spending the last few weeks testing local music generation on Apple Silicon, mostly around ACE-Step 1.5 + MLX.

Reddit

Ingenious? Orwellian? Or both? Supreme Court considers constitutionality of 'geofence' warrants

Reddit

Meta’s $2 billion Manus acquisition blocked by China.

Reddit

Qwen3.6-27B-3bit-mlx · Hugging Face: 3 & 5 mixed quant for RAM poor Mac users.

Reddit