Benchmarks / Long Context & Instruction
AA Long Context Reasoning
Artificial Analysis's long-context reasoning benchmark.
maintained by Artificial Analysis · 220 observations on record
Versions
1 version
| Version | Tasks |
|---|---|
| AA-LCR current aa-lcr | — |
Current leaderboard
AA-LCR · Accuracy · top 25 of 220 systems
| # | System | Accuracy % |
|---|---|---|
| 1 | GPT 5.2 Codexxhigh effort OpenAI | 75.7% |
| 2 | GPT 5 (2025-08-07)high effort OpenAI | 75.6% |
| 3 | GPT 5.1 (2025-11-13)high effort OpenAI | 75.0% |
| 4 | GPT 5.5 (2026-04-22)xhigh effort OpenAI | 74.3% |
| 5 | GPT 5.6 Lunamax effort OpenAI | 74.0% |
| 6 | Claude Opus 4.5thinking Anthropic | 74.0% |
| 7 | Minimax M3 MiniMax | 74.0% |
| 8 | GPT 5.3 Codexxhigh effort OpenAI | 74.0% |
| 9 | GPT 5.4 (2026-03-05)xhigh effort OpenAI | 74.0% |
| 10 | GPT 5.6 Terramax effort OpenAI | 74.0% |
| 11 | GPT 5.6 Solmax effort OpenAI | 73.7% |
| 12 | GPT 5.5 (2026-04-22)high effort OpenAI | 73.3% |
| 13 | Mimo V2 5 Pro xiaomi | 73.3% |
| 14 | GPT 5 (2025-08-07)medium effort OpenAI | 72.8% |
| 15 | GPT 5.2 (2025-12-11)xhigh effort OpenAI | 72.7% |
| 16 | Gemini 3.1 Pro Preview Google DeepMind | 72.7% |
| 17 | GPT 5.6 Terrahigh effort OpenAI | 72.3% |
| 18 | GPT 5.5 (2026-04-22)medium effort OpenAI | 72.3% |
| 19 | GPT 5.5 (2026-04-22)low effort OpenAI | 72.0% |
| 20 | GPT 5.6 Terraxhigh effort OpenAI | 71.3% |
| 21 | GLM 5.2max effort Z.ai | 71.3% |
| 22 | Gemini 3.5 Flashmedium effort Google DeepMind | 71.0% |
| 23 | GPT 5.6 Solxhigh effort OpenAI | 71.0% |
| 24 | Claude Sonnet 4.6max effort Anthropic | 70.7% |
| 25 | Claude Sonnet 5max effort Anthropic | 70.7% |
One row per evaluated system — reasoning-effort variants rank separately.