An Artificial Analysis benchmark that simulates knowledge workers such as analysts, researchers, and lawyers reasoning over large document collections.
Browse the latest scores, model modes, release dates, and parameter sizes for AA-LCR.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
 Hy3 Thinking Level · High | 73.40 | 2026-07-06 | 295B | Free Commercial | |
 Gemini 3.0 Pro (Preview 11-2025) Thinking Level · High | 71.00 | 2025-11-18 | Unknown | Closed | |
 Claude Sonnet 4.6 Thinking Enabled | 71.00 | 2026-02-17 | Unknown | Closed | |
4 |  MiniMax M2.5 Thinking Enabled | 69.50 | 2026-02-12 | 229B | Free Commercial |
5 | MiniMax-M2.7 Thinking EnabledTools | 69.00 | 2026-03-18 | 229B | Non-Commercial |
6 |  Qwen3.5-397B-A17B Thinking Enabled | 68.70 | 2026-02-16 | 39.7B | Free Commercial |
7 | Qwen 3.6 Plus Preview Thinking Enabled | 68.30 | 2026-03-31 | Unknown | Closed |
8 | Qwen3.5-27B Thinking Enabled | 66.10 | 2026-02-25 | 27B | Free Commercial |
9 | Gemini 2.5-Pro Thinking Enabled | 66.00 | 2025-06-05 | Unknown | Closed |
10 | Claude Sonnet 4.5 Thinking Enabled | 66.00 | 2025-09-30 | Unknown | Closed |
11 | Claude Sonnet 4 Thinking Enabled | 65.00 | 2025-05-23 | Unknown | Closed |
12 |  Kimi K2.5 Thinking Enabled | 65.00 | 2026-01-27 | 1000B | Free Commercial |
13 |  GLM-5 Thinking Enabled | 63.00 | 2026-02-11 | 744B | Free Commercial |
14 | Claude Sonnet 3.7 Thinking Enabled | 61.00 | 2025-02-25 | Unknown | Closed |
