Morgan(@morganlinton):I said that Qwen 27B can't be used for production-grade code, and a bunch of people told me I was wrong. And I might be wrong. That was my assumption, but probably silly for me to make an assumption that I haven't validated myself. So I'm bumping up Qwen 27B to be the next model I benchmark with @VulcanBench. Some very reliable people have told me it's a very good model, so my bad for speaking before benchmarking. Let's let the data do the talking!

Morgan

@morganlinton

cofounder @boldmetrics / studying how llms think @vulcanbench / early @sonos views are my own

Joined January 2009

1.2K Following 47.9K Followers

Morgan@morganlinton

2026.08.18 17:20

I said that Qwen 27B can't be used for production-grade code, and a bunch of people told me I was wrong. And I might be wrong. That was my assumption, but probably silly for me to make an assumption that I haven't validated myself. So I'm bumping up Qwen 27B to be the next model I benchmark with @VulcanBench. Some very reliable people have told me it's a very good model, so my bad for speaking before benchmarking. Let's let the data do the talking!

287

Forward to community

Most Popular Users