reasoning benchmark
FrontierCode Leaderboard
FrontierCode is Cognition's coding evaluation of models on difficult coding tasks and production codebase standards.
Updated Sep 5, 2026
Models4
Model coverage4
MetricScore
EvidenceB
FrontierCode Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelXA | Score61.30% | Percentile100.00% | Participants4 | EvidenceC | Evaluated |
| Rank02 | ModelAN | Score53.40% | Percentile66.67% | Participants4 | EvidenceC | Evaluated |
| Rank03 | ModelAN | Score46.30% | Percentile33.33% | Participants4 | EvidenceC | Evaluated |
| Rank04 | ModelAN | Score38.80% | Percentile0.00% | Participants4 | EvidenceC | Evaluated |
FrontierCode Highlights
The leading models and scores on this benchmark.
FrontierCode Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for FrontierCode
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is FrontierCode?
What FrontierCode measures and how its scores work.
FrontierCode is a coding evaluation by Cognition, with a Diamond subset containing the hardest problems.
It measures Score as a ratio for performance on difficult coding tasks and adherence to production codebase standards.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- FrontierCode
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about FrontierCode.
Which model scores highest on FrontierCode?
Grok 4.6 is currently ranked first with 61.30%.
What are the top three models on FrontierCode?
The current leaders are Grok 4.6 (61.30%), Claude Opus 5 (53.40%), and Claude Fable 5 (46.30%).
Which FrontierCode model has the lowest official input price?
Grok 4.6 has the lowest matched official input price at $2.0 input / $6.0 output per 1M tokens.
Which models are fastest among FrontierCode results?
The fastest matched records are Grok 4.6 (132.69 tok/s via xAI), Claude Opus 5 (62.88 tok/s via Anthropic), and Claude Fable 5 (43.18 tok/s via Anthropic).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.