Medical Long Context Reasoning (MLCR-AA)
Professional Work
Artificial Analysis' private held-out expert and compound subset of Wisedocs Medical Long Context Reasoning. Responses must be concise, nonempty, accurate and complete by majority vote of three judges; three attempts per question are averaged.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5.1 | 71.1 |
| Claude Fable 5 | 64.4 |
| Claude Opus 5 | 59.4 |
| Sonnet 5 | 55.0 |
| GLM-5.3 Flash | 51.1 |
| GLM-5.3 | 48.3 |
| Opus 4.8 | 45.6 |
| Muse Spark 1.3 | 43.3 |
| Kimi K3 | 38.3 |
| GPT-6 Astra | 35.0 |
| GPT-5.6 Terra | 31.7 |
| Muse Spark 1.2 | 31.1 |
| Opus 4.7 | 28.9 |
| GPT-5.6 Sol | 26.1 |
| Sonnet 4.6 | 24.4 |
| DeepSeek V4.1 Flash | 22.8 |
| Gemini 3.8 Flash | 21.7 |
| Qwen3.8 27B | 21.7 |
| DeepSeek-V4-Pro | 21.1 |
| Muse Glimmer 30B | 20.0 |
| Qwen3.8 Max (0902) | 20.0 |
| GPT-5.6 Luna | 19.4 |
| Qwen3.8-Max | 19.4 |
| Gemini 3.5 Flash | 18.3 |
| DeepSeek V4 Pro 0813 | 17.8 |
| GPT-5.5 | 17.8 |
| MiniMax M3 | 17.2 |
| Step 5 Preview | 16.7 |
| GPT-6 Luna | 16.1 |
| GPT-6 Sol | 16.1 |
| Kimi K2.7 Code | 16.1 |
| Gemini 3.1 Pro Preview | 15.6 |
| DeepSeek-V4-Flash | 15.0 |
| Gemini 3.7 Flash | 15.0 |
| Grok 4.5 | 15.0 |
| Grok 4.7 | 15.0 |
| Gemini 3.6 Flash | 14.4 |
| DeepSeek-V4-Flash-0731 | 13.3 |
| Grok 4.6 | 12.2 |
| Inkling | 12.2 |