AA Long Context Reasoning: leaderboard
AA-LCR: ability to extract, reason about, and synthesize information from documents of 10k-100k tokens (100 questions). Proprietary AA benchmark.
Metric: Accuracy (%). Source: artificialanalysis.ai. Status: saturation imminent. 509 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K3 (Max) | 88.67 |
| 2 | Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) | 85.33 |
| 3 | Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback) | 84.67 |
| 4 | GPT-5.5 (xHigh) | 84.33 |
| 5 | GPT-5.5 (High) | 84.33 |
| 6 | Muse Spark 1.3 (Max) | 84.33 |
| 7 | GPT-5.6 Sol (Max) | 84 |
| 8 | Gemini 3.8 Flash (Medium) | 84 |
| 9 | GPT-5.6 Luna (Max) | 83.67 |
| 10 | Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) | 83.67 |
| 11 | Muse Glimmer 30B (High) | 83.33 |
| 12 | GPT-5.3 Codex (xHigh) | 83.33 |
| 13 | MiniMax-M3 | 83 |
| 14 | GPT-5.6 Terra (Max) | 83 |
| 15 | Muse Spark 1.3 (xHigh) | 83 |
Interactive version: theaggregate.ai/benchmark?slug=aa-long-context-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.