PhysAssistBench (Chinese): leaderboard
The Chinese version of PhysAssistBench (2026), a benchmark of LLMs as physician assistants: over 324 four-turn sessions built from MIMIC-IV records, the model answers underspecified physician requests by querying an EHR through FHIR tools and interviewing an agentic patient. Scored by the mean share of physician-validated rubric items passed, in percent; higher is better.
Metric: Mean rubric score (%, mRS) over the 1,296 Chinese turns (324 four-turn clinical sessions built from MIMIC-IV records: 4 scenarios by 3 data-richness tiers by 27 sessions); the model assists a physician with 18 FHIR EHR, patient-interview and control tools (at most 16 tool calls per turn) and each turn scores the fraction of binary rubric items a fixed GPT-5.4-mini judge marks passed; thinking enabled (GPT-5 series at high effort), temperature 0.2; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5 | 71.5 | #260 |
| 2 | Kimi K2.6 | 69.9 | #178 |
| 3 | Claude Opus 4.7 (Thinking) | 69.9 | #79 (Claude Opus 4.7) |
| 4 | Seed 1.8 | 68.7 | #242 |
| 5 | Gemini 3.1 Pro (Preview) | 68 | #94 |
| 6 | MiniMax-M2.7 | 66.8 | #518 |
| 7 | GPT-5.4 (High) | 65.9 | #123 (GPT-5.4) |
| 8 | Qwen 3.5 35B A3B | 65.8 | #482 |
| 9 | GPT-5.4 Mini (High) | 61.5 | #389 (GPT-5.4 Mini) |
| 10 | Qwen 3.5 27B | 61.4 | #359 |
| 11 | Qwen 3.5 9B | 58 | #803 |
| 12 | Qwen 3.5 4B | 47.4 | #1175 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=physassistbench-chinese · How It Works · Data refreshed daily, snapshot 2026-11-13.