Featured
AI video generation
New 2026
ComfyUI node for unattended MiniMax H3 prompt-chain video generation
New 2026
AI skill for diagnosing, structuring, writing, and evaluating Chinese lead-generation short-video scripts.
New 2026
All-in-one MiniMax H3 node for ComfyUI - T2V, I2V, R2V, audio drive, keyframes, extend & Motion Context chains in a single node.
New 2026
Maestro AI port to AMD GPU
New 2026
ComfyUI node for MiniMax H3 that chains multiple video clips with motion context, disk caching, dynamic image references, audio reference support, and seamless final video/audio decoding.
New 2026
Generative video re-detailer for ComfyUI. Runs your clip back through LTX-2.5's IC-LoRA pixel spatial upscaler: a drag-and-drop workflow plus a CLI that handles sizing, cut-based chunking, frame arithmetic and audio. It synthesizes detail rather than restoring it.
New 2026
A topic goes in, a finished film comes out — an all-round AI filmmaking agent that runs script, shot breakdown, character sheets, image, voice and final cut on your own machine.
New 2026
Local-first Windows workflow for turning gameplay recordings into polished vertical clips.
New 2026
MiniMax-H3 video+audio generation on AMD Strix Halo (gfx1151) via ComfyUI + ROCm 7.14 — working setup, benchmarks, and a fix for corrupt published weights
New 2026
DreamX-Phi 1.0: Action-Conditioned Video World Model for Robotic Manipulation
New 2026
inference code for minimax h3
New 2026
Vertical STEM explainer videos, built from a validated spec instead of generated animation code
New 2026
Every prompt from the official Wan 3.0 creator handbook, each paired with the official result clip — 30s takes, multi-shot timecodes, reference-driven and audio-synced cases.
19125
New 2026
A hardware-aware Codex skill for local MiniMax H3 video generation through ComfyUI
New 2026
Codex skill for creating vertical Chinese book-list videos with hand-drawn illustrations, Doubao TTS and bilingual subtitles.
New 2026
Open-source local implementation of MiniMax H3 Context-IR. Turn simple prompts into structured, validated H3 video briefs.
New 2026
Streaming, steerable text 2 video: a Wan2.1-1.3B student distilled with SF DMD that generates video continuously instead of as a fixed clip.
Other topics
Browse other topics on Trendshift
A
N