FULL STORY
Qwen 3.8 Leak and the 35B Mystery
A GitHub commit revealed traces of Alibaba's Qwen 3.8 with 35B total and 3B active parameters. A follow-up commit removed the model, suggesting the 35B size may never ship.
2026-08-15 ~ 2026-08-18 · 2 episodes · 4 posts
Episode 1 · Alibaba's Qwen 3.8 Model Parameters Leaked (2026-08-15, 2 posts)
Code commits reveal a potential Qwen 3.8 model with 35B total parameters and 3B active parameters, featuring MoE architecture and aggressive memory optimization.
- Qwen 3.8 35BA3B model spotted in GitHub commit — BazzyIm · 2026-08-15
- Qwen3.8-35B-A3B Spotted in GitHub Code, Optimized for Consumer GPUs — cephaloform · 2026-08-15
Episode 2 · Qwen 35B model appears scrapped as code commits remove it (2026-08-16, 2 posts)
Code commits suggest Qwen's 35B model has been removed and may never ship, while a Qwen developer hinted users shouldn't wait for the 35B-A3B variant, fueling speculation about its replacement.
- Commits suggest Qwen 35B model removed, likely not releasing — Local-Cardiologist-5 · 2026-08-16
- Qwen Dev Hints: Don't Wait for the 35B-A3B Model — Mean-Ad1493 · 2026-08-18