alibaba AI models

All 12 alibaba models tracked on Reviuws, with pricing, context length, licensing and community reviews.

  • Qwen2.5-VL-72B — alibaba: Qwen2.5-VL-72B provides advanced visual understanding including document parsing, video comprehension and object grounding. It supports a 128k context window an
  • Qwen3.8-Omni-Flash — alibaba: Qwen3.8-Omni-Flash is Alibaba's cutting-edge omni-modal AI model designed to seamlessly process text, image, and audio inputs simultaneously. Built specifically
  • Qwen3-235B-A22B — alibaba: Qwen3-235B-A22B is a mixture-of-experts model with 235B total/22B active parameters, supporting seamless switching between thinking and non-thinking modes. It s
  • Qwen 3 235B — alibaba: Alibaba's Qwen 3 235B is an exceptionally powerful open-weight model optimized for elite multilingual understanding and advanced programming tasks. Built on a m
  • Qwen3.5-397B-A17B — alibaba: Alibaba's Qwen3.5-397B-A17B is the debut model of the Qwen3.5 family, leveraging a massive Mixture of Experts architecture with 397 billion total and 17 billion
  • Qwen3.8-27B — alibaba: Dense 27B open-weight Qwen3.8 release for self-hosting and fine-tuning.
  • Qwen3.8-Flash-Next — alibaba: Multimodal 125B MoE with just 6B active parameters per token — an early preview of the Qwen4 architecture, open-weighted with FP8 checkpoint.
  • Qwen3-Max — alibaba: Alibaba's largest Qwen release, a trillion-parameter-scale mixture-of-experts model with a production thinking mode for coding, reasoning and agentic tasks. It
  • Qwen3.5-397B-A17B — alibaba: Alibaba's first Qwen3.5 release: 397B-parameter MoE with 17B active parameters, Apache 2.0.
  • Qwen-Image-2.1 — alibaba: Open image generation and editing model with transparent-image output and support for multiple visual references.
  • Qwen-Audio 3.1 Realtime Plus — alibaba: Realtime full-duplex voice model with function calling, voice customization and long-session context.
  • Qwen3.8-Max — alibaba: Alibaba's Qwen3.8-Max is a groundbreaking 2.4-trillion parameter vision-language model utilizing a highly efficient Mixture-of-Experts architecture with 95 bill