HF RL Explorer

I want loaded AutoencoderKL objects to provide a pure latent-to-image decode API: decode(z…

I want loaded AutoencoderKL objects to provide a pure latent-to-image decode API: decode(z…: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). I want loaded AutoencoderKL objects to provide a pure latent-to-image decode API: decode(z: torch.FloatTensor, return dict: bool =…

The task

I want loaded AutoencoderKL objects to provide a pure latent-to-image decode API: decode(z: torch.FloatTensor, return_dict: bool = True) -> AutoencoderKLOutput. The method should take a batch of latent tensors, optionally apply the VAE post-quantization projection, run the…

Part of XiaomiMiMo/MiMo-V2.6-RL-oss.