phi-4-reasoning-plus tuned for scalable inference with long context using Unsloth.
Want to make some of these yourself?
