Phi 4
[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion…
Shogo Agent Index
UnrankedNot ranked yet: 0 agentic benchmarks measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.
Agentic coding
13
estimated
Long-horizon autonomy
23
estimated
Knowledge-work agents
13
estimated
Computer use
3
estimated
Security & context learning
11
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 41 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 3 | |
| DeepSWE | est. | 6% | predicted 5 | |
| FrontierSWE | est. | 5% | predicted 0 | |
| CursorBench | est. | 4% | predicted 13 | |
| GSO-Bench | est. | 4% | predicted 1 | |
| METR Time Horizons | est. | 10% | predicted 42 | |
| Vending-Bench 2 | est. | 6% | predicted 1 | |
| PostTrainBench | est. | 4% | predicted 9 | |
| APEX-Agents | est. | 8% | predicted 6 | |
| GDPval | est. | 7% | predicted 12 | |
| Remote Labor Index | est. | 5% | predicted 0 | |
| DeepResearch Bench | est. | 5% | predicted 39 | |
| OSWorld 2.0 | est. | 6% | predicted 0 | |
| OSWorld | est. | 4% | predicted 8 | |
| Cybench | est. | 5% | predicted 11 | |
| CL-bench | est. | 5% | predicted 11 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| BALROG | Long-horizon | 11.6% | source |
| CritPt | Reasoning | 0.0% | source |
| GPQA Diamond | Reasoning | 56.1% | source |
| MATH Level 5 | Math | 64.9% | source |
| MMLU | Legacy | 84.8% | source |
| OTIS Mock AIME | Math | 13.8% | source |