GLM-5.3-FlashX: Launches on B.AI Platform
GLM-5.3-FlashX hits 200 tokens/sec on B.AI via 320B sparse model with 1M context and native multimodal support.
SourceAnalysis
GLM-5.3-FlashX is now live on B.AI API and Web Chat, a speed-optimized native multimodal model from Z.AI built on a 320B total / 18B active parameter sparse architecture that delivers up to 200 tokens/sec, a 1M context window, and native text/image/video/file understanding for high-speed coding and agent workflows.
Justin Sun 孙宇晨
@justinsuntronJustin Sun is the founder of TRON, BitTorrent ($BTT) owner and crypto exchange HTX advisor
Premium Sponsors
.jpg)