Asked Qwen 3.8 A3B (MoE) mixture of experts model to create an interactive island in 3D with three js.

After working for 15 minutes while spitting out 10 tokens per second it generated this cute floating island. Not very bad for a locally run AI Model! 😎LM Studio Bionic harness

Husni MarwanNov 61

Here is proof that Qwen 3.8 27B can interact with a DAW and compose music.

Cool, Alibaba_Qwen!

Dr. Tristan BehrensNov 62

Concept Country has developed its first open-source model, Qwen Concept 27B.

With 54% less thinking tokens and 3% better performance on GPQA Diamond compared to its base Qwen 3.8 model, we have achieved 30% more useful tokens per joule without additional post-training.

ConceptNov 6123

Testing many models, many engines, many quants.

These are two exceptional outputs from everyone's favorite local model, Qwen 3.8 Flash. In all models, certain weights matter more than others. Some engines are fast because they round off the important bits. I think that's why

Mooncast ProductionsNov 63