o3-mini
Smaller o-series reasoner for economical coding, math, and planning tasks
Shogo Agent Index
Estimated#85 of 9223.3range 14.9–31.6 · 9% of weight measured
Agentic coding
24
11% measured
Long-horizon autonomy
34
estimated
Knowledge-work agents
21
estimated
Computer use
13
estimated
Security & context learning
19
50% measured
Price
- Input
- $0.99/M
- Output
- $4.00/M
- Cache read
- —/M
- 3:1 blend
- $1.74/M
- Agent-effective
- $1.13/M
Agent-effective = 20 input tokens per output token, 85% cache hits. Why.
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 59 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 19 | |
| DeepSWE | est. | 6% | predicted 16 | |
| FrontierSWE | est. | 5% | predicted 2 | |
| CursorBench | est. | 4% | predicted 22 | |
| GSO-Bench | 1.3% | 4% | OpenHands | |
| METR Time Horizons | est. | 10% | predicted 58 | |
| Vending-Bench 2 | est. | 6% | predicted 6 | |
| PostTrainBench | est. | 4% | predicted 14 | |
| APEX-Agents | est. | 8% | predicted 17 | |
| GDPval | est. | 7% | predicted 23 | |
| Remote Labor Index | est. | 5% | predicted 1 | |
| DeepResearch Bench | est. | 5% | predicted 43 | |
| OSWorld 2.0 | est. | 6% | predicted 1 | |
| OSWorld | est. | 4% | predicted 29 | |
| Cybench | 22.5% | 5% | source | |
| CL-bench | est. | 5% | predicted 15 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| Aider Polyglot | Coding | 60.4% | source |
| ARC-AGI | Reasoning | 34.5% | source |
| ARC-AGI-2 | Reasoning | 3.0% | source |
| Chess Puzzles | Reasoning | 15.0% | source |
| CritPt | Reasoning | 0.3% | source |
| DTBench | Knowledge | 68.8% | source |
| Fiction.LiveBench | Long context | 50.0% | source |
| FrontierMath Tiers 1-3 | Math | 18.6% | source |
| FrontierMath Tier 4 | Math | 0.0% | source |
| GPQA Diamond | Reasoning | 77.0% | source |
| LMCA | Knowledge | 19.0% | source |
| MATH Level 5 | Math | 96.5% | source |
| OTIS Mock AIME | Math | 76.9% | source |
| SciCode | Coding | 42.8% | source |
| SimpleBench | Reasoning | 22.8% | source |
| SimpleQA Verified | Knowledge | 15.3% | source |