vLLM-Omni adds day-0 support for MiniMax H3-Image on Vera Rubin GPUs
AIvLLM-Omni says it brought MiniMax H3-Image to NVIDIA Vera Rubin, with 0.87-second median 4K generation on four GPUs in a short-prompt serving benchmark. A separate throughput configuration reaches 6,115 images per hour with four single-GPU replicas.
Source details
vLLM Blog · vLLM-Omni team · vllm.aiPublished · added here
Original title: H3-Image Day-0 Support: 4K in 0.87s with Vera Rubin