← sources

Minimax H3 generation time for different Nvidia GPUs

source: reddit (reddit:comfyui)published 2026-09-17 14:03ingested 2026-09-18 02:41

retained text

I am new to comfyui, my current setup is dual 7900xtx + 32 GB ddr5 5600 that I built to run Qwen 3.6 series models because they were the cheapest per GB. Lately I started playing with comfyui, minimax H3 is amazing. 

Now I am looking to build a rig specifically for image/video generation, and based on the posts I read here, Nvidia should be the go to GPU. Thinking about getting a 3090/5070ti or may be a 4090, If I can find one. But I have no idea on how fast Nvidia GPUs are relative to the AMD e.g 7900xtx. People post the numbers in many threads but I am not sure about their workflow and what optimization they have done.

  
When running below comfyui included  T2V template.

https://preview.redd.it/pegum36n53qh1.png?width=283&format=png&auto=webp&s=45d2aebac7c420ddcf8b34e7df673d7dc883a4ff

My generation time is:  
5 Seconds video at 0.4 MP: **5 min 27s - 11.66s/it (20 iteration)**

5 Seconds video at 1.0 MP: **15 min 52s - 45.77s/it** **(20 iteration)**

Here are the models used in workflow (May be I am not using correct model for my setup?)

https://preview.redd.it/c4zrb8yt63qh1.png?width=627&format=png&auto=webp&s=56836fbb88bc80e50041180d00ea2348b4b48647

Could you guys help me with generation time number on you GPU with above template? without any lora preferably, so 20 iterations.

  
Just trying to understand the gap and which Nvidia card is best bang for buck.s

document fetched 2026-09-18 00:42 · canonical https://reddit.com/r/comfyui/comments/1wiuwyn/minimax_h3_generation_time_for_different_nvidia