Microsoft Launched Several New “Open” Ai Models On Wednsday, the most capable of which is competitive with Openai’s O3-Min on at Least One Benchmark.
All of the new pemissively licensed models-phi 4 mini reasoning, phi 4 reasoning, and phi 4 reasoning plus-are “Reasoning” models, meaning they’re able to spend more time more time more conducts toom founds to complex proposals. They expand microsoft’s phi “small model” family, which the company launched a year ago out of a foundation for ai developers bulding apps at the edge.
Phi 4 mini reasoning was trained on roughly 1 Million synthetic Math Problems Generated by Chinese AI Startup Deepsek’s R1 Reasoning Model. Around 3.8 billion parameters in size, phi 4 mini reasoning is designed for educational applications, microsoft says, like “Embedded Tutoring” on Lightweight Devices.
Parameters roughly correspond to a model’s problem-saolving skills, and models with more parameters generally performance better than there, than those with fewer parameters.
Phi 4 reasoning, a 14-billion-paarameter model, was trained using “High-Quality” Web Data as Well as “Curated Demonstrations” from Openai’s AFOREMENTIOND O3-Mi. It’s Best for Math, Science, and Coding Applications, According to Microsoft.
As for Phi 4 Reasoning Plus, It’s Microsoft’s Previous-Released Phi-4 Model Adapted Into a Reasoning Model to achieve better according to participation on Particular Tasks. Microsoft Claims that Phi 4 Reasoning Plus Approaches The Performance Levels of R1, a model with significantly more parameters (671 billion). The company’s internal benchmarking also has phi 4 reasoning plus matching o3-mini on omnimath, a Math skills test.
Phi 4 mini reasoning, phi 4 reasoning, and phi 4 reasoning plus are available on the AI Dev Platform Hugging Face Accompanied by Detailed Technical Reports.
Techcrunch event
Berkeley, Ca
,
June 5
“Using distillation, reinforcement learning, and high-quality data, these [new] Models balance size and performance, “Wrote Microsoft in A blog post“They are small enough for low-latency environments. Tasks efficiently. “