← All benchmarks
RCM-Bench-1: RCM knowledge
Health & Medicine · 2026-10-02
Pass rate over three attempts on each of the 48 RCM-Bench-1 scenarios for answering revenue cycle management knowledge questions.
Benchmark source: Commure
Top models (higher is better)
| Model | Score | Source |
|---|
| Gemini 3.8 Flash | 97.9 | RCM-Bench-1 results: Gemini 3.8 Flash, RCM knowledge · Commure |
| Claude Opus 5 | 96.5 | RCM-Bench-1 results: Claude Opus 5, RCM knowledge · Commure |
| Claude Opus 5.5 | 95.8 | RCM-Bench-1 results: Claude Opus 5.5, RCM knowledge · Commure |
| GPT-5.6 Sol | 95.8 | RCM-Bench-1 results: GPT-5.6 Sol, RCM knowledge · Commure |
| GPT-6.1 Sol | 95.8 | RCM-Bench-1 results: GPT-6.1 Sol, RCM knowledge · Commure |
| Grok 4.7 | 95.8 | RCM-Bench-1 results: Grok 4.7, RCM knowledge · Commure |
| Gemini 4 Argon | 95.1 | RCM-Bench-1 results: Gemini 4 Argon, RCM knowledge · Commure |
| Sonnet 5 | 93.8 | RCM-Bench-1 results: Claude Sonnet 5, RCM knowledge · Commure |
| GPT-6 Sol | 93.8 | RCM-Bench-1 results: GPT-6 Sol, RCM knowledge · Commure |
| GPT-5.6 Terra | 93.1 | RCM-Bench-1 results: GPT-5.6 Terra, RCM knowledge · Commure |
| Claude Sonnet 5.5 | 92.4 | RCM-Bench-1 results: Claude Sonnet 5.5, RCM knowledge · Commure |
| GPT-6 Luna | 91.7 | RCM-Bench-1 results: GPT-6 Luna, RCM knowledge · Commure |
| GPT-5 Mini | 90.3 | RCM-Bench-1 results: GPT-5 mini, RCM knowledge · Commure |
| Claude Haiku 5.5 | 89.6 | RCM-Bench-1 results: Claude Haiku 5.5, RCM knowledge · Commure |
| Sonnet 4.6 | 87.5 | RCM-Bench-1 results: Claude Sonnet 4.6, RCM knowledge · Commure |
| DeepSeek V4.1 Flash | 86.1 | RCM-Bench-1 results: DeepSeek V4.1 Flash, RCM knowledge · Commure |
| GLM-5.3 | 86.1 | RCM-Bench-1 results: GLM-5.3, RCM knowledge · Commure |
| GPT-5.6 Luna | 84.7 | RCM-Bench-1 results: GPT-5.6 Luna, RCM knowledge · Commure |
| Opus 4.6 | 81.3 | RCM-Bench-1 results: Claude Opus 4.6, RCM knowledge · Commure |
| Haiku 4.5 | 56.3 | RCM-Bench-1 results: Claude Haiku 4.5, RCM knowledge · Commure |