Completely satisfied Birthday, Phi! To rejoice, we’re introducing three new fashions within the Phi-4 reasoning household on Azure AI Foundry:
1️⃣Phi-4-reasoning (14B, SFT on curated reasoning demos)
2️⃣Phi-4-reasoning-plus (14B, SFT + RL, ~1.5x inference-time tokens)
3️⃣Phi-4-mini-reasoning (3.8B, distilled + SFT with DeepSeek-R1-generated math issues)
These fashions leverage distillation, SFT, RL, and artificial knowledge pipelines to realize frontier-level functionality in compact kind. They’re optimized for multi-step reasoning, mathematical downside fixing, and inside reflection—duties historically dominated by a lot bigger frontier fashions.
Information right here: https://lnkd.in/gC_56Dfy