codingmoh 2 days ago

I saw pretty good reasoning quality with phi-4-mini. But alright - I’ll still run some tests with qwen2.5-coder and plan to add support for it next. Would be great to compare them side by side in practical shell tasks. Thanks so much for the pointer!