reasoning benchmark
FrontierCode 1.1
FrontierCode 1.1 evaluates whether coding-agent changes are mergeable, using unit tests, maintainer-defined rubrics, and verifiers. Runs flagged for unfair internet use receive a zero score.
Updated Aug 7, 2026
Published models15
Registry coverage15
MetricScore
EvidenceB
FrontierCode 1.1 leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AN | 53.5% | 100.0% | 15 | B | |
| 2 | AN | 53.4% | 92.9% | 15 | B | |
| 3 | OP | 47.5% | 85.7% | 15 | B | |
| 4 | AN | 46.5% | 78.6% | 15 | B | |
| 5 | OP | 43.0% | 71.4% | 15 | B | |
| 6 | AN | 42.7% | 64.3% | 15 | B | |
| 7 | XA | 42.4% | 57.1% | 15 | B | |
| 8 | OP | 41.3% | 50.0% | 15 | B | |
| 9 | OP | 39.8% | 42.9% | 15 | B | |
| 10 | AN | 38.5% | 35.7% | 15 | B | |
| 11 | MA | 30.1% | 28.6% | 15 | B | |
| 12 | ZA | 24.5% | 21.4% | 15 | B | |
| 13 | DE | 17.6% | 14.3% | 15 | B | |
| 14 | MI | 14.7% | 7.1% | 15 | B | |
| 15 | AC | 10.2% | 0.0% | 15 | B |
Score distribution
Top published rows on the benchmark's original scale.
FrontierCode 1.1 highlights
The top published results on this benchmark's own scale.
What is FrontierCode 1.1?
Definition and scoring fields from the benchmark registry.
FrontierCode 1.1 evaluates whether coding-agent changes are mergeable, using unit tests, maintainer-defined rubrics, and verifiers. Runs flagged for unfair internet use receive a zero score.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- FrontierCode 1.1
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- frontiercode-1.1|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about FrontierCode 1.1.
Which model scores highest on FrontierCode 1.1?
Claude Fable 5 is currently ranked first with 53.5%.
What does FrontierCode 1.1 measure?
FrontierCode 1.1 evaluates whether coding-agent changes are mergeable, using unit tests, maintainer-defined rubrics, and verifiers. Runs flagged for unfair internet use receive a zero score.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
15 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is marked as eligible for the current LLMBoard capability methodology.