MiniMax H3 community shares local generation workflows and tips
Posts show MiniMax H3 generating video locally on GPUs from RTX 3060 to 5070 Ti, with one user reporting about 2 minutes of generation per second of footage. A creator said they deleted WAN 2.2, claiming H3 handles out of the box what WAN needed LoRAs for. Fizgig LoRA trainer v4.1.2 adds training on video clips and voice recordings, per community round-ups.
How this story unfolded
4 days · 0 reports · 58 community posts · from Aug 16
- Aug 16
- Aug 17
- Aug 18
- Aug 19
- Aug 20
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Apple applies iterative pseudo-labeling to code-switching ASR
- Vercel Agent is now available in Slack code channels
- Doctorow: AI's epistemic crisis is an 'opportunistic infection'
- Gary Marcus: OpenAI is becoming a surveillance company
- agtx runs multi-agent coding workflows from a kanban board