minimax h3

2 articles · newest first

Night line-art stopwatch next to the headline: One AI video clip, app, API or a rented GPU? A card reads: Wan 2.2, 5 s clip, 24 GB GPU, 71 s with the Lightning LoRA.

AI video cost per clip: an app, an API, or a rented GPU? (And what's legal in the US)

I read the Reddit threads, the licences and the price pages so you don't have to. What one 5-second clip really costs in Higgsfield, through an API, and on a rented RTX 4090, plus the MiniMax H3 licence line almost nobody mentions.

Ari Vale · 3h ago · 14 min

Night line-art GPU card next to the headline: The cheapest GPU for ComfyUI is a 24 GB card. A price card reads RTX 3090 on Vast, on-demand, from $0.12 an hour.

Cheapest cloud GPU for ComfyUI: images and video (Flux, Wan 2.2, MiniMax H3)

A 24 GB card runs Flux and Wan 2.2 at 640x640, and you never need an H100 for it. Which GPU each workflow needs, today's prices on Vast and Runpod, the out-of-memory fixes people actually use, and the setup that doesn't waste your first hour.

Ari Vale · Sep 23 · 8 min