Alibaba ATH's video generation model HappyHorse-1.0: API test status on Model Studio, open weights availability, Mac local inference reality, and which GPU to pick on RunPod.
Local video generation test on M1 Max 64GB MacBook Pro. FP8 models don't work on Metal — switching to GGUF got Wan 2.2 running at 82 minutes for a 2-second clip. LTX-2 produced NaN or unusable KSampler output under MPS. Specs, failed configs, and the working setup.
