Baidu: ernie-4.5-vl-28b-a3b-pt:free
A powerful multimodal Mixture-of-Experts chat model featuring 28B total parameters with 3B activated per token, delivering exceptional text and vision understanding through its innovative heterogeneous MoE structure with modality-isolated routing. Built with scaling-efficient infrastructure for high-throughput training and inference, the model leverages advanced post-training techniques including SFT, DPO, and UPO for optimized performance, while supporting an impressive 131K context length and RLVR alignment for superior cross-modal reasoning and generation capabilities.
Todos los proveedores de este modelo están ocupados ahora mismo
Cada proveedor ascendente alcanzó su límite de velocidad. El modelo vuelve automáticamente cuando los límites se levantan, normalmente en cuestión de horas. Inténtalo de nuevo en un rato o cambia a otro modelo.
Solicitar este modelo en Discord