long context benchmark
AA-LCR
Agent Arena Long Context Reasoning benchmark
Updated Aug 11, 2026
Models16
Model coverage16
MetricScore
EvidenceB
AA-LCR Ranking
Higher score ranks better on this benchmark.
| 1 | ME | 80.0% | 100.0% | 16 | C | |
| 2 | TE | 73.4% | 93.3% | 16 | C | |
| 3 | MA | 71.2% | 86.7% | 16 | C | |
| 4 | MA | 70.0% | 80.0% | 16 | C | |
| 5 | AC | 68.7% | 73.3% | 16 | C | |
| 6 | AC | 68.3% | 66.7% | 16 | C | |
| 7 | AC | 66.9% | 60.0% | 16 | C | |
| 8 | AC | 66.1% | 53.3% | 16 | C | |
| 9 | NV | 65.4% | 46.7% | 16 | C | |
| 10 | AC | 63.0% | 40.0% | 16 | C | |
| 11 | MI | 62.0% | 33.3% | 16 | C | |
| 12 | AC | 58.5% | 26.7% | 16 | C | |
| 13 | NV | 58.3% | 20.0% | 16 | C | |
| 14 | AC | 57.0% | 13.3% | 16 | C | |
| 15 | AC | 25.6% | 6.7% | 16 | C | |
| 16 | AC | 4.7% | 0.0% | 16 | C |
AA-LCR Score Distribution
A closer view of the leading scores on this benchmark.
AA-LCR Highlights
The leading models and scores on this benchmark.
What is AA-LCR?
What AA-LCR measures and how its scores work.
Agent Arena Long Context Reasoning benchmark
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- AA-LCR
- Modality
- text
- Primary category
- long context
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- aa-lcr|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about AA-LCR.
Which model scores highest on AA-LCR?
Muse Glimmer-30B is currently ranked first with 80.0%.
What does AA-LCR measure?
Agent Arena Long Context Reasoning benchmark
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
16 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.