Microsoft’s most succesful new Phi 4 AI mannequin rivals the efficiency of far bigger methods

Microsoft on Wednesday launched a number of new “open” AI fashions, probably the most able to which is aggressive with OpenAI’s o3-mini on at the least one benchmark.

Because it says on the tin, the entire new permissively licensed fashions — Phi 4 mini reasoning, Phi 4 reasoning, and Phi 4 reasoning plus — are “reasoning” fashions, that means they’ll spend extra time fact-checking options to advanced issues. They increase Microsoft’s Phi “small mannequin” household, which the corporate launched a yr in the past to supply a basis for AI builders constructing apps on the edge.

Phi 4 mini reasoning was educated on roughly 1 million artificial math issues generated by Chinese language AI startup DeepSeek’s R1 reasoning mannequin. Round 3.8 billion parameters in measurement, Phi 4 mini reasoning is designed for academic purposes, Microsoft says, like “embedded tutoring” on light-weight gadgets.

Parameters roughly correspond to a mannequin’s problem-solving expertise, and fashions with extra parameters typically carry out higher than these with fewer parameters.

Phi 4 reasoning, a 14-billion-parameter mannequin, was educated utilizing “high-quality” internet information in addition to “curated demonstrations” from OpenAI’s aforementioned o3-mini. It’s finest for math, science and coding purposes, in accordance with Microsoft.

As for Phi 4 reasoning plus, it’s Microsoft’s previously-released Phi 4 mannequin tailored right into a reasoning mannequin to realize higher accuracy for specific duties. Microsoft claims Phi 4 reasoning plus approaches the efficiency ranges of DeepSeek R1, which has considerably extra parameters (671 billion). The corporate’s inner benchmarking additionally has Phi 4 reasoning plus matching o3-mini on OmniMath, a math expertise check.

Phi 4 mini reasoning, Phi 4 reasoning, Phi 4 reasoning plus, and their detailed technical experiences, can be found on the AI dev platform Hugging Face.

“Utilizing distillation, reinforcement studying, and high-quality information, these [new] fashions steadiness measurement and efficiency,” wrote Microsoft in a weblog put up. “They’re sufficiently small for low-latency environments but preserve robust reasoning capabilities that rival a lot greater fashions. This mix permits even resource-limited gadgets to carry out advanced reasoning duties effectively.”