ToolNavs Find Useful AI Tools
Submit Sign in
Back to AI News Briefing
Nov 16 AI Briefing: Qwen4-Max Launches, Tencent's Hy4 Flash Debuts, IMF Warns on AI Financing Risks

Nov 16 AI Briefing: Qwen4-Max Launches, Tencent's Hy4 Flash Debuts, IMF Warns on AI Financing Risks

AI News Briefing • Admin • • 15 views

On November 16, the AI world saw two major model launches in a single day: Alibaba released Qwen4-Max and Tencent rolled out the lower-cost Hy4 Flash. On the same day, the IMF's managing director warned that circular financing in AI has topped $1 trillion, while OpenAI launched a science blog with a physics lesson from GPT-6 Astra. Here are today's four stories:

Qwen4-Max launches: 2.4 trillion parameters, 62% lower task costs

Alibaba released Qwen4-Max on November 16, a model with 2.4 trillion parameters, 71 billion activated per token, a 1-million-token context window and native vision. The company says it outperforms the previous Qwen3.8-Max across coding, agentic tasks and visual reasoning, and that average cost per task on its Work360 benchmark fell from $4.43 to $1.67, a 62% drop. The model is live in Qwen apps and via API at $1.50 per million input tokens and $4.50 per million output tokens; Alibaba says smaller Qwen4 models will follow as open weights.

Tencent's Hy4 Flash goes live: one-twentieth the task cost of Hy4

Tencent launched Hy4 Flash the same day, a lower-cost model aimed at everyday work: 224 billion total parameters, 12 billion activated per token, and the same million-token context. API pricing is $0.07 per million input tokens and $0.281 per million output tokens, roughly one-twelfth of flagship Hy4; on Tencent's internal WorkIndex benchmark, cost per task is one-twentieth of Hy4's. The model is available in WorkBuddy, Yuanbao and Tencent Cloud TokenHub; WorkBuddy is free until November 30 and TokenHub API calls are half price, and Tencent plans to open-source the weights.

IMF chief warns: the global economy is "standing on one leg" on AI

IMF Managing Director Kristalina Georgieva warned on November 16, in a CNBC interview on the sidelines of the ASEAN summit, that the world economy is "standing on one leg" on AI investment. The IMF estimates circular financing among AI companies has exceeded $1 trillion, about two-thirds of it funded by debt. She said a sudden loss of investor confidence in AI and a withdrawal of capital could drag down global growth and create financial stability problems, and that the IMF is closely monitoring the concentration risk.

OpenAI launches a science blog: GPT-6 Astra joins a Standard Model proof

OpenAI launched a science blog on November 15. Its first guest post comes from UC Santa Barbara physicist Nathaniel Craig, who worked with GPT-6 Astra on a candidate lattice definition of the Standard Model, aiming to close a foundational gap that has troubled physics for more than 50 years. Craig stresses it is only "an answer in principle", not a practical tool: the key property of reflection positivity remains unresolved and the paper has not yet been peer-reviewed; the underlying preprint was posted to arXiv on October 6.

Recommended Tools

More