Minimax H3 test: rock eating rocks, 20-min generation on RTX PRO 6000
Moarkush · reddit · 2026-08-08
A Reddit user shares a test of Minimax H3 text-to-video model, generating a clip of 'rock eating rocks' in 20 minutes on an RTX PRO 6000. The user is mostly pleased but notes some jelly-like deformation when the rock is picked up.
More from Multimodal
- Krea2 Users Report Lack of Quality Custom Checkpoints, Recommend Base Model with LoRAs — More_Bid_2197 · 2026-08-08
- Local Video Generation: MiniMax Lags Far Behind LTX in Inference Speed — PhilosopherSweaty826 · 2026-08-08
- Micro-expressions Matter More Than Pixel Count in AI Video — aftahi_ai · 2026-08-08
- Runway Launches Seedance 2.5: 30-Second Videos with Sound from 50 References — JeffSynthesized · 2026-08-08
- Seedance 2.5 generates realistic UGC videos, accelerating brand ad creative automation — Aiden_Tech_Ai · 2026-08-08
- MiniMax H3 vs. Seedance 2.0: A Side-by-Side Video Generation Test — ajrss2009 · 2026-08-08