I want to find someone who is obsessed with local models

I’m fucking totally obsessed.

Not much. Running gpt-oss:20b via Ollama completely offline. For personal stuff I never use remote models.

i've experimented a bit, lm studio, vllm, on a ryzen 5 ai + 96GB system ram, the problem isn't really the inference, it's the prefill that's super slow

I've tried aider-chat and claude code with different local models. It worked but it took like an hour to do what I asked to do. nostr:nevent1qqsqqqzfdena73k92cwp8d3p3ucsq6dmw04a4u8pcqrg33wgq6823ncpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhsygy7xr55qguvm847h33js9md6ngsnqfp99zz72nv8pe8l3n05l4fpgpsgqqqqqqsfe...

https://www.amd.com/en/developer/resources/technical-articles/2026/how-to-run-a-one-trillion-parameter-llm-locally-an-amd.html

What do u want to know