FRAMEWIREIndonesiaUpdated Aug 16Live wire
0:00 / 0:00

"The first example is snow globe, look at how good Qwen3.8-27b result is, it got the refraction correct

We can see full refraction + reflection, i know its glass, but glimmer just gave up." So is a 30B dense better than 27B dense? Surely since 30 > 27 right? btw, you can skip

AJAug 169
0:00 / 0:00

Running Qwen 3.8 27B Q4 on a single RTX 4060 (8GB VRAM) gaming laptop shouldn’t be physically possible.

You can now run this 27B dense beast locally with a massive 64,000 token context window (minimum requirement by hermes agent) on budget consumer hardware, without crashing

AlokAug 16150
0:00 / 0:00

Alibaba’s Qwen team just released a genuinely usable open model, not another oversized preview nobody can run.

Qwen 3.8 27B, Apache 2.0 license — fully open, commercial use included. 27B parameters with vision built in, 262K native context extensible to 1M. The FP8 variant

AI PulseAug 16
0:00 / 0:00

Qwen 3.8 27b one-shot meteor game with 4bit unsloth quant so it works comfortably in 24GB RTX 4090.

I was trying to answer the question: "is qwen 3.8 good enough for local coding?" I didn't focus on tok/s because you can hack that with small contexts. My demo was against

Julian HarrisAug 163