Issues

• Maximum VRAM during diffusion based model inference

  nvidia-cosmos/cosmos-predict1#145 · Filter by author LDS666888 opened on Jan 4, 2026

• Upgrade or avoid pinning old huggingface-hub

  nvidia-cosmos/cosmos-predict1#144 · Filter by author hongbo-miao opened on Jan 1, 2026
  1

• Can not run tokenizer on CPU or MPS devices

  nvidia-cosmos/cosmos-predict1#143 · Filter by author AmitMY opened on Oct 28, 2025

• Multi-Resolution Text2World Training

  nvidia-cosmos/cosmos-predict1#142 · Filter by author YuxuanSnow opened on Aug 27, 2025

• Please release Medusa-fine-tuned 4B?

  nvidia-cosmos/cosmos-predict1#140 · Filter by author lenn-arts opened on Aug 7, 2025

• About the metric Latent L2

  nvidia-cosmos/cosmos-predict1#133 · Filter by author B0B8K1ng opened on Jul 10, 2025

• Can smaller number of frames be generated as output?

  P1question
  nvidia-cosmos/cosmos-predict1#120 · Filter by author nnsriram27 opened on Jun 5, 2025
  1