Alibaba releases Qwen4-Max flagship model, 2.4 trillion parameters
Start here
Alibaba officially released the Qwen4-Max flagship model, adopting the new architecture of the Qwen4 series, with total parameters of 2.4 trillion and 71B activated per token, featuring native vision and a million-token context window. It has been launched on the Qianwen AI platform to provide API services. The model surpasses models such as Claude Opus5.5 and GPT-6.1 Sol on Agent-specific benchmarks like DeepSWE and AA-Briefcase. Work360 evaluation shows its single-task cost is reduced by 58% compared to the previous generation Qwen3.8-Max. Currently, Qwen4-Max cache hit costs as low as 0.5 yuan per million tokens. Users can access it via Alibaba Token Plan, or experience it in applications such as Qianwen Office, Qoder, Qianwen APP, and Qwen Studio; Qwen4 series small and medium-sized models will be open-sourced and released successively.
Written by AI from the reports · updated 3 days ago
Report timeline
Follow the reports to see the event from other sides.
- 智东西Alibaba releases Qwen4-Max flagship model, 2.4 trillion parameters, single-task cost down 58%
Alibaba officially released the Qwen4-Max flagship model, adopting the new architecture of the Qwen4 series, with total parameters of 2.4 trillion and 71B activated per token, featuring native vision and a million-token context window. The model surpasses models such as Claude Opus5.5 and GPT-6.1 Sol on Agent-specific benchmarks like DeepSWE and AA-Briefcase. Work360 evaluation shows single-task cost reduced by 58% compared to the previous generation Qwen3.8-Max. It has been launched on the Qianwen AI platform to provide API services.
Heat trend for this event
- Current
- 3
- Peak
- 10Nov 16 14:00
- 24h change
- -51%
The trend compares only sources observed the whole time, so the range may be smaller than the current heat figure. Hover or tap the chart for each hour. Left and right arrow keys move through it.