AA-LCR leaderboard
As of November 6, 2026, among the results listed here, Sakana Fugu Max has the highest reported AA-LCR score, 83, published by Sakana, ahead of Reflection Beam 1.0 501B A23B at 79.4. 9 models have a AA-LCR result in our records.
Scores as the model makers published them, not re-run by us; each row links to its announcement.
9 models ranked · 7 makers
| Rank | Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|---|
| 1 | Fugu Max | 83 | Not published | Sakana | June 24, 2026 (model release) · read from the maker's chart |
| 2 | R Beam 1.0 501B A23B | 79.4 | FP8 | Reflection | October 16, 2026 (model release) |
| 3 | Kimi K3 | 74.7 | max | Moonshot AI | July 16, 2026 (model release) |
| 4 | Hy3 | 73.4 | Not published | Tencent | July 6, 2026 (model release) · read from the maker's chart |
| 5 | Solar Pro 4 | 71 | Not published | Upstage | August 10, 2026 (model release) |
| 6 | Hy3 preview | 66.3 | Not published | Tencent | July 6, 2026 (model release) · read from the maker's chart |
| 7 | Ling 3.0 Tiny | 58.7 | Thinking | inclusionAI | August 6, 2026 (model release) · read from the maker's chart |
| 8 | Nemotron 3 Super | 58.4 | Not published | NVIDIA | August 11, 2026 (model release) |
| 9 | Nemotron 3.5 Lightning | 52 | Not published | NVIDIA | August 11, 2026 (model release) |
Other reported runs
| Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|
| R Beam 1.0 501B A23B | 78.2 | BF16 | Reflection | October 16, 2026 (model release) |
| R Beam 1.0 501B A23B | 78.7 | NVFP4 | Reflection | October 16, 2026 (model release) |
How much to trust these numbers
- Source
- Reported by the model makers themselves. Not re-run by LLMBenchmark or anyone independent.
- Settings
- Settings differ between makers. Check the Setting column before comparing two scores.
- Spread
- 83 to 52 across 9 models.
- Freshness
- Newest result published October 16, 2026; oldest June 24, 2026.
- Charts
- Some values were read from charts in the announcements. Two independent readings had to match exactly.
What AA-LCR measures
AA-LCR asks models to answer 100 questions requiring reasoning across real-world document sets averaging 100,000 tokens; the search results do not explain what its scores mean. Source: ArtificialAnalysis/AA-LCR · Datasets at Hugging Face