BlogLLM Benchmark Scores as Published, Side by Side
MedXpertQA: Results as Published
Every result the catalog holds for MedXpertQA: 18 models from 6 developers, read from 4 pages. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area HealthcareMaintainer Tsinghua University (Yuxin Zuo and co-authors)Introduced 2025Data version 2026103001, as of 2026-10-30
MedXpertQA tests expert-level medical knowledge and reasoning with 4,460 exam-style questions across 17 specialties and 11 body systems. It has a text subset and a multimodal (MM) subset whose questions come with medical images and rich clinical information. Scores are the share of questions answered correctly.
MedXpertQA MM
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.2OpenAI, released 2025-12-11 | 73.3% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| GPT-5 miniOpenAI, released 2025-08-07 | 34.4% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Claude Opus 4.5Anthropic, released 2025-11-24 | 63.6% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Claude Sonnet 4.5Anthropic, released 2025-09-29 | 54% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Gemma 4 12BGoogle, released 2026-04-02 | 48.7% | Google, Gemma 4 model cardchecked 2026-10-15 | - |
| Gemma 4 26B A4BGoogle, released 2026-04-02 | 58.1% | Google, Gemma 4 model cardchecked 2026-10-15 | - |
| Gemma 4 31BGoogle, released 2026-04-02 | 61.3% | Google, Gemma 4 model cardchecked 2026-10-15 | - |
| Gemma 4 E2BGoogle, released 2026-04-02 | 23.5% | Google, Gemma 4 model cardchecked 2026-10-15 | - |
| Gemma 4 E4BGoogle, released 2026-04-02 | 28.7% | Google, Gemma 4 model cardchecked 2026-10-15 | - |
| Gemini 3 ProGoogle, released 2025-11-18 | 76% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Qwen3.5-122B-A10BAlibaba, released 2026-02-24 | 67.3% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Qwen3.5-27BAlibaba, released 2026-02-24 | 62.4% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Qwen3.5-35B-A3BAlibaba, released 2026-02-24 | 61.4% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Qwen3.5-397B-A17BAlibaba, released 2026-02-16 | 70% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Qwen3-VL-235B-A22BAlibaba, released 2025-09-23 | 47.6% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 1 more page | - |
| Kimi K2.5Moonshot AI, released 2026-01-27 | 65.3% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
MedXpertQA Text (accuracy)
accuracy
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Claude Sonnet 4.6Anthropic, released 2026-02-17 | 49% | Microsoft, Microsoft AI: MAI-Thinking-1 technical reportchecked 2026-10-19 | Effort: max |
| MAI-Thinking-1Microsoft, released 2026-08-12 | 43% | Microsoft, Microsoft AI: MAI-Thinking-1 technical reportchecked 2026-10-19 | - |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.