AISystem

Qwen3

Alibaba's 2025 open-weight Qwen3 language-model family, the workload brought up on Redwood in its third week and the model that then found optimizations for its own operations on the chip.

open-weight-model

Sources

Attribution and reporting linked by this record; links do not imply independent verification.

Qwen3 is the 2025 generation of Alibaba’s Qwen model family, released with a technical report on arXiv (paper); later members of the family in this corpus include Qwen3-Max-Thinking and Qwen3.7-Max. Here it is the workload Architect Labs brought online on Redwood in the accelerator’s third week and demonstrated live at the Design Automation Conference. Exposed as an endpoint inside the AI system that built the chip, it found timing and kernel optimizations for its own operations — the moment the newsletter calls the loop reaching silicon.

Development timeline

Newest reporting dates first. Reporting dates may differ from event dates.

    Knowledge graphopen in browser →

    Developed by

    Cite or correct this record

    Suggest a correction — opens a GitHub draft for you to review and submit.