Tracked topic
… My local stack keeps growing I think this one earns a permanent spot, though not by pushing anything else out. Qwen and Gemma are still doing the bulk of what I need from a local model. …
… One of my tests, though, proved just how incredible local models have become. There are reasons to believe the hype when it comes to the Qwen models; I've had consistently good experiences with Qwen 3.6 27B , and Qwen 3.8 27B is, so far, more of the same but better. …
… In one test, with 90,000 tokens of context and zero contamination, the local model gave the better answer. This isn't local models beating the cloud, because they don't overall . It's messier than that, but Qwen gives it some fantastic competition. …
… Related I ditched LM Studio for an open-source alternative — and my local model is doing things it couldn't before It's better in all the ways I needed local AI to be better A quick overview of Qwen and Gemma Why these two ended up in my rotation Qwen 3.5 9B dropped in February as part of the Qwen … …
… Related I finally found a local LLM I want to use every day and it's not for coding Local AI that actually fits into my day Qwen 3.5 4B My unplanned third pick Phi-3.5 mini was supposed to be the third model. …
… I suspect that most will bring their own cloud-based subscription, but I was especially curious how a local LLM would handle it instead. Armed with Qwen 3.6 27B Q4 K M with multi-token prediction on a Radeon RX 7900 XTX and deployed with llama.cpp, I was ready to go. …