Public Artificial Intelligence Benchmark Records Database
Phi-4
MODEL RECORD: MSFT-PHI4
Table may be scrolled horizontally on narrow displays.
| Record No. | Model | Benchmark | Metric | Score | Eval Date | Source | Model ID |
|---|---|---|---|---|---|---|---|
BR-027PHI4-001 |
Phi-4 | Artificial Analysis Intelligence Index | Overall | 4.9 | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-002 |
Phi-4 | GPQA Diamond | Accuracy | 57.5% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-003 |
Phi-4 | Humanity's Last Exam No tools | Overall | 4.1% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-004 |
Phi-4 | SciCode | Overall | 26% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-005 |
Phi-4 | tau2-bench | Overall | 0% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-006 |
Phi-4 | Terminal-Bench Hard | Overall | 3.8% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-007 |
Phi-4 | CritPt | Overall | 0% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-008 |
Phi-4 | AA-Omniscience | Overall | -56.7 | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-009 |
Phi-4 | AA-Omniscience | Accuracy | 13.2% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-010 |
Phi-4 | IFBench | Overall | 23.5% | Unknown | Primary source | MSFT-PHI4 |
BR-027PHI4-011 |
Phi-4 | AA-LCR | Overall | 0% | Unknown | Primary source | MSFT-PHI4 |
Showing records 1–11 of 11
Page 1 of 1