Records found: 4
'Qwen3-Max-Preview is Alibaba's first trillion-parameter LLM, offering a 262K-token context window and competitive benchmark performance while remaining API-only with tiered pricing.'
'Alibaba's Ovis2.5 (9B and 2B) advances multimodal AI with a native-resolution vision transformer and an optional thinking mode, achieving top scores for open-source models under 40B and improved OCR and chart understanding.'
Alibaba introduces Lumos-1, a unified autoregressive video generation model leveraging MM-RoPE and AR-DF to efficiently model spatiotemporal dependencies, achieving competitive results with efficient training.
WINGS introduces a dual-learner architecture that balances attention between text and images in multimodal large language models, significantly reducing text-only forgetting and boosting performance across benchmarks.