Benchmark
AA-LCR
Run by Artificial Analysis rather than the vendor. Third-party execution is the point, but it also means the vendor did not control the setup.
- Released
- 2025-09-16
- Categories
- long_context
- Openness
- unknown
- Source
- model_reports
- Reported scores
- 2
Reported scores
model_reports
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Kimi K3 | Moonshot AI | 74.7 | 2026-06-13 | evidence |
| Qwen3.5-397B-A17B | Qwen | 68.7 | 2026-02-16 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.