Loving Qwen 3.8 27B for text animations in HTML canvas.
Here is a side-by-side between Qwen 3.8 27B (left) and Claude Opus 4.8 (right) Same audio, generated by Opus 4.8, but the canvas rendering is unique to each model with the same prompt. I actually think I can pretty
We live in a new world.
Unlimited, free super intelligence on our desks Qwen 3.8 27B locally on a RTX 5090 32 GB VRAM system with 115 tokens/sec What do you think? 😎@Alibaba_Qwen
Gemini 3.7 Flash, Qwen 3.8, Grok 4.6, GLM 5.3.
Four models, one race. Which of these have you actually tested and which one surprised you?
This is unreal. Qwen 3.8 27b running locally on my RTX 5090 BEATS Opus 4.8 in my benchmarks
(funny enough slightly loses to Opus 4.6 lol) Speeds are just as good if not sometimes MUCH better. Got up to 200 tk/s in some tests We live in a new world. Unlimited, free super
You have reached the end of the archive
All of qwen38