Pretty fast 👀 Qwen 3.8 125B / Mixed 4bit-8bit.
NEW THINGS FOR NeiroWoof BROPUMP 👊⛽️
Let me show you the actual running results of Lazymao AI computing power warehouse with Qwen 3.8 27B 786K
Sometimes the code output is so fast that the screen refreshes. The 786K context does not affect the speed. Instead, it makes the memory for writing code 3 times larger than the official Qwen one. There are not many machines with discounts on Lazymao AI computing power cabin. If you want this kind of effect, buy it quickly. If you hesitate again, the price increase will be crazy after a while 😂…
I tried using Qwen 3.8 27B
I can do this in one shot on my crappy spec PC. I would have been quite surprised if it was cloud AI from a while ago.
Qwen 3.8 Max has arrived — but does it really deliver all of this?
👀 In today's video, I reacted to the SWEN benchmark and commented on the performance of the new model, comparing the results and raising an important question: to what extent do the benchmarks really show the ability to
You have reached the end of the archive
All of qwen38