Low-Spec Guide: Running MiniMax H3 Video Generation on 8GB VRAM
No_Operation7634 · reddit · 2026-08-13
A developer shared a detailed workflow for running the MiniMax H3 video generation model on low-VRAM hardware (RTX 4060 8GB + 16GB RAM).
- Performance: Achieves 5 seconds of 0.2 MP (608x352) video generation in just 1 minute 35 seconds. Resolution can be increased if video duration is reduced.
- Configurations: Uses 20-step Euler Simple sampler, specific attention models, and swaps in a lightweight text encoder (like qwen3vl4bint8) to save memory.
- Open Source: The author shared complete FL2V and REF2V workflow configurations for the community to reuse.
Related event: MiniMax H3 Video Generation Tested on Low-End Hardware(3 posts)→
More from Infra
- New hardware day: local LLM deployment the moment it arrives — lxfater · 2026-10-16
- Together AI's score centering fixes RL training collapse under quantized inference — nrehiew_ · 2026-10-16
- Inference is splitting prefill and decode across specialized chips, not just H100s — AccBalanced · 2026-10-16
- Rebasely launches in beta: Git hosting built for humans and coding agents — handotdev · 2026-10-16
- Drama3 TTS lands in audio.cpp: up to 5x real-time on RTX 5090, <8GB VRAM mode too — Acceptable-Cycle4645 · 2026-10-16
- Three CoreWeave co-founders have sold $3.3B in shares since IPO amid credit worries — SumitGup · 2026-10-16