WTF guys, why is literally nobody yapping about qwen 3.8 flash rn?!
Tested the exact same one-shot flappy bird prompt, and the speed-cost-output balance is actually insane solid 8/10. yes some barrier placement is kinda weird, and GLM-5.3 flash has slightly smoother visuals,
Your phone got power with superintelligent
Locally Qwen 3.8 27B on a 3090, over Tailscale. agent designing a camera app with Live preview, capture, filters, zoom, vision for object detection, 155 tok/s prompt, 51 tok/s generate,
This is my version of the same prompt made by Qwen 3.8 Q4_K_XL reasoning low, on 2x3090 24gb.
It took 13 mins to complete, and 36k tokens context.
Companies are moving AI in-house for control and privacy, fueling growth in local AI options.
Qwen 3.8 27B now offers unprecedented local intelligence, a game-changer for software development.
You have reached the end of the archive
All of qwen38