Alibaba's Qwen3.7-Plus supports text, video and imagery inputs at low cost of $0.4/$1.6 per 1M token — but it's proprietary
Alibaba recently launched Qwen3.7-Plus, a new AI language model that features multimodal capabilities and is 60% cheaper than its predecessor, Qwen3.7-Max, but is only available under a closed commercial license. This model aims to replace expensive frontier models in enterprise workflows by offering high performance in visual tasks and maintaining operational continuity through advanced context management features, despite concerns about compliance and the lack of open-source access.
For enterprises considering the integration of AI models, Qwen3.7-Plus by Alibaba offers a compelling case due to its cost-effective multimodal capabilities and advanced context management features. The model's ability to handle high-frequency developer workflows and its compatibility with OpenAI's infrastructure make it a viable alternative to more expensive models like GPT-5, especially for tasks involving visual workflows and repetitive operations. However, the closed commercial licensing and reliance on Alibaba Cloud for model deployment pose compliance challenges for organizations with strict data residency requirements.