Skip to content
Agientry
← Back to Videos

FastH3 Reply Test: About One-Quarter the Render Time

Loading video
Loading video

In a reply under the quoted FastH3 V2 post, @cySG20 shares a 10-second vertical fire-tailed dragon clip and reports roughly one-quarter the render time with little quality loss.

Video GenerationFastH3 OmniFastH3 V2MiniMax H3VgenvAccelerated InferenceGenerated Video
Category: video-generation
Author: @eternityspring
Date: 2026-10-21T01:09:01
Duration: 10.208s
Source: https://x.com/eternityspring/status/2112712730155311227
Reference: https://huggingface.co/collections/FastVideo/fasth3-omni

Related Videos

FastH3 Omni cuts a 15-second 480p clip to about 230 seconds
0:15

FastH3 Omni cuts a 15-second 480p clip to about 230 seconds

@hk_nick · 24 views · 2026-10-21
Video GenerationFastH3 Omni
FastVideo open-sources FastH3 Omni: 12x faster generation on NVIDIA Blackwell, training code included
0:28

FastVideo open-sources FastH3 Omni: 12x faster generation on NVIDIA Blackwell, training code included

@haoailab · 98 views · 2026-10-20
Video GenerationFastH3 Omni
Qwen-Image and MiniMax H3 Generate a Short Drama
1:05

Qwen-Image and MiniMax H3 Generate a Short Drama

@eternityspring · 17 views · 2026-10-21
MiniMax H3Qwen-Image
MiniMax H3 Character-Swap LoRA Preserves Motion
0:06

MiniMax H3 Character-Swap LoRA Preserves Motion

@IsekaiBardGuild · 60 views · 2026-10-22
Video GenerationCharacter Swap
FastH3 V2 on an RTX 4080: 4.4x Faster Than Official MiniMax H3
0:08

FastH3 V2 on an RTX 4080: 4.4x Faster Than Official MiniMax H3

@eternityspring · 90 views · 2026-10-07
Video GenerationFastH3
Character-Swap LoRA v1.1: An 80 MB Likeness LoRA for MiniMax H3 in ComfyUI
0:19

Character-Swap LoRA v1.1: An 80 MB Likeness LoRA for MiniMax H3 in ComfyUI

@Alissonerdx · 145 views · 2026-10-18
Video GenerationCharacter Swap

As an Amazon Associate, we earn from qualifying purchases.