Alibaba Cloud Model Studio has released Qwen4-Max, the new flagship model in the Qwen family, with support for a 1,000,000 token context window and multimodal inputs. While it is designed to be compatible with the OpenAI Chat Completions protocol, a specific feature for encrypted reasoning does not function correctly when integrated with the Vercel AI SDK. This issue arises because the SDK drops the encrypted reasoning field, preventing the model from retaining its prior thought process across conversation turns. A workaround involves using the SDK's metadata extractor to preserve and resend this encrypted data. AI
IMPACT This release introduces a large context window and multimodal capabilities, but highlights potential integration challenges with existing SDKs for advanced features like encrypted reasoning.
RANK_REASON New flagship model release from a major AI lab (Alibaba Cloud Model Studio). [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →