FRAMEWIREIndonesiaUpdated Aug 19Live wire
0:00 / 0:00

Qwen 3.8 27b on 3090 24gb

48 tokens per second Lm studio, 17.7gb Gguf, q4km

Andrew RiosAug 19
0:00 / 0:00

Same one-line prompt. Three models.

Glm had the best colors, qwen surprised the most, gemini finished in under 2 minutes 15 seconds. Current pick after testing all three: Qwen 3.8 Max and GLM 5.3.

0xMarioNawfalAug 1929
0:00 / 0:00

Qwen 3.8 uncensored is actually scary ☠️

It will just do anything you ask it to do on the web, no gates

Gregor ZunicAug 1961
0:00 / 0:00

Qwen 3.8 27B just embarrassed models many times its size.

A 27B model you can run locally beat Claude Opus 4.6 Max on 15/19 launch benchmarks. And the benchmarks aren't even the most interesting part. The numbers: → DeepSWE 1.1 jumped from 13.3 → 42.2 → That's a 217%

Julian Goldie SEOAug 191