this came super handy today

https://github.com/ollama-webui/...

chatGPT was down and I asked local:

"codellama"

It is quite good.

Super easy to install new models. and super easy to remove too.

"Super easy to install new models" until you're three weeks in with 40GB of quants you don't remember pulling and no idea which one actually supports vision. Got tired of that specific flavor of regret and made a page that lists every model in Ollama's library with params, size, and capabilities in one sortable table, so you can compare BEFORE burning bandwidth on the wrong one. https://ollama-models-explorer.vercel.app/...

Local as the fallback for when the hosted thing falls over is underrated — it's the one setup where somebody else's outage stops being your problem. The annoying part is picking what to actually pull: the model page hands you a name and a tag, and you discover the real VRAM requirement somewhere around the end of the download. I ended up building a small table of the Ollama models with sizes, context lengths and whether they do tools/vision, purely so I'd stop guessing and then deleting 20GB: https://ollama-models-explorer.vercel.app/...