- SWE-Bench Verified 71.6
- SWE-Bench Pro 51.2
- SWE-Bench Multilingual 65.5
- Terminal Bench 2 54.8
- AIME 2026 92.5
- AMO Bench 40
- Frontier Math (Tier 1-3) 6.3
- HLE (Academic reasoning, text) 18
- GPQA Diamond 84.6
- Frontier Science 58.2
- Artifacts Bench 36.4
- IF Bench (average) 75
- Advanced IF (average) 71.4
- Robust IF Bench (internal diverse hard IF, average) 61.2
- 𝜏²-Bench telecom 71.7
MAI-Code-1-Flash
MAI-Code-1-Flash.
15
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Microsoft Jun 2026 MAI-Code-1-Flash
Mid-size
No Epoch Capabilities Index score yet
- IFBench 75% #23 /70
- ArtifactsBench 36.4% #5 /5
- SWE-bench Multilingual 65.5% #51 /70
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| IFBench | 75% | #23 of 70 | Model card | |
| Humanity's Last Exam | 18% | #107 of 147 | Model card | |
| AdvancedIF | 71.4% | #6 of 6 | Model card | |
| Robust IF Bench | 61.2% | #1 of 1 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 84.6% | #75 of 284 | Model card | |
| AIME 2026 | 92.5% | #23 of 34 | Model card | |
| AMO-Bench | 40% | #3 of 4 | Model card | |
| FrontierScience | 58.2% | #6 of 7 | Model card | |
| FrontierMath | 6.30% | #11 of 11 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 71.6% | #66 of 150 | Model card | |
| SWE-bench Multilingual | 65.5% | #51 of 70 | Model card | |
| SWE-Bench Pro | 51.2% | #63 of 77 | Model card | |
| ArtifactsBench | 36.4% | #5 of 5 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Terminal-Bench 2.0 | 54.8% | #31 of 61 | Model card | |
| tau2-bench | 71.7% | #52 of 69 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it.
MAI-Code-1-Flash model card 2 Jun 2026
Capabilities
- SWE-bench Verified 71.6
- SWE-Bench Pro 51.2
- SWE-bench Multilingual 65.5
- Terminal-Bench 2.0 54.8
- AIME 2026 92.5
- AMO-Bench 40
- FrontierMath 6.3
- Humanity's Last Exam 18
- GPQA Diamond 84.6
- FrontierScience 58.2
- ArtifactsBench 36.4
- IFBench 75
- AdvancedIF 71.4
- Robust IF Bench 61.2
- tau2-bench 71.7