Fully Local Animation with MiniMax H3 on an RTX 5070Ti
maciusik · reddit · 2026-09-28
A Reddit user shares a fully local animation pipeline: the Plaguekind-modified MiniMax-H3 (up to 9 reference images to video, weights on Hugging Face) with Qwen 3.6 Heretic for prompts, running on an RTX 5070Ti, 32GB RAM and an i5 CPU.
More from Infra
- Why Doesn't Nvidia Just Sell Tokens Itself? Ex-Nvidia Engineer: "Jensen Makes His Friends Billionaires" — rohanpaul_ai · 2026-09-28
- Berkeley's TRACE robots trace monochrome data center cables via bidirectional tracing — berkeley_ai · 2026-09-28
- Brookings paper: US AI buildout to cost $10.3 trillion through 2032, 3.6% of GDP yearly — alex_verem · 2026-09-28
- Kokoro-82M: tiny open-source TTS model runs on CPU with quality of far larger models — solyarisoftware · 2026-09-28
- India's Sarvam AI can now host American AI models on its own infrastructure — rvp · 2026-09-28
- MoEspresso runs Qwen3.8-Flash-Next (125B) at 12-15 tok/s on a 2021 32GB M1 Max — marcobaldo · 2026-09-28