Benchmarks · long-context
MLCR-AA
Medical Long Context Reasoning (MLCR-AA): reasoning over long medical records; overall score.
69 models · 141 results · data as of 2026-10-29 · artificialanalysis.ai ↗
Leaderboard
Sonnet 5.5 max
All results
141 rows| # | Model | Effort | MLCR-AA |
|---|---|---|---|
| 1 | Sonnet 5.5 | max | 41.7% |
| 2 | GPT-6 Astra | max | 41.7% |
| 3 | GPT-6 Astra | medium | 38.3% |
| 4 | GPT-6.1 Sol | max | 38.3% |
| 5 | GPT-6 Astra | xhigh | 36.1% |
| 6 | GPT-6.1 Sol | xhigh | 35.6% |
| 7 | Haiku 5.5 | max | 35.0% |
| 8 | GPT-6 Astra | low | 33.9% |
| 9 | GPT-6 Sol | max | 33.9% |
| 10 | GPT-6 Astra | high | 32.8% |
| 11 | GPT-6 Sol | xhigh | 32.2% |
| 12 | GPT-6.1 Sol | medium | 32.2% |
| 13 | GPT-6.1 Sol | high | 31.7% |
| 14 | GPT-6.1 Terra | max | 30.6% |
| 15 | Qwen 3.8 2.4T-A95B | 30.6% | |
| 16 | GPT-5.6 Sol | max | 30.0% |
| 17 | GPT-6 Sol | high | 29.4% |
| 18 | GPT-5.5 | xhigh | 28.9% |
| 19 | GPT-6 Sol | medium | 28.9% |
| 20 | Opus 5.5 | max | 27.8% |
| 21 | GPT-5.6 Sol | xhigh | 27.8% |
| 22 | GPT-5.6 Terra | max | 27.2% |
| 23 | DeepSeek V4.1 Flash 1017 | max | 26.7% |
| 24 | GPT-6.1 Sol | low | 26.7% |
| 25 | GPT-5.6 Sol | medium | 26.1% |
| 26 | GPT-6 Luna | max | 26.1% |
| 27 | GPT-6 Sol | low | 25.6% |
| 28 | Opus 5.5 | xhigh | 25.0% |
| 29 | GPT-5.5 | high | 25.0% |
| 30 | GPT-5.6 Sol | high | 24.4% |