Atlas

Benchmarks

← All benchmarks

RCM-Bench-1: Insurance updates

Health & Medicine · 2026-10-02

Pass rate over three attempts on each of the 5 RCM-Bench-1 scenarios for updating coverage, mapping payers and relinking insurance to claims.

Benchmark source: Commure

Top models (higher is better)

ModelScoreSource
Sonnet 593.3RCM-Bench-1 results: Claude Sonnet 5, Insurance updates · Commure
Gemini 4 Argon93.3RCM-Bench-1 results: Gemini 4 Argon, Insurance updates · Commure
Claude Haiku 5.580.0RCM-Bench-1 results: Claude Haiku 5.5, Insurance updates · Commure
Opus 4.680.0RCM-Bench-1 results: Claude Opus 4.6, Insurance updates · Commure
Claude Opus 580.0RCM-Bench-1 results: Claude Opus 5, Insurance updates · Commure
Sonnet 4.680.0RCM-Bench-1 results: Claude Sonnet 4.6, Insurance updates · Commure
Gemini 3.8 Flash80.0RCM-Bench-1 results: Gemini 3.8 Flash, Insurance updates · Commure
GLM-5.380.0RCM-Bench-1 results: GLM-5.3, Insurance updates · Commure
GPT-5.6 Sol80.0RCM-Bench-1 results: GPT-5.6 Sol, Insurance updates · Commure
GPT-5.6 Terra80.0RCM-Bench-1 results: GPT-5.6 Terra, Insurance updates · Commure
GPT-5 Mini80.0RCM-Bench-1 results: GPT-5 mini, Insurance updates · Commure
GPT-6.1 Sol80.0RCM-Bench-1 results: GPT-6.1 Sol, Insurance updates · Commure
Claude Opus 5.573.3RCM-Bench-1 results: Claude Opus 5.5, Insurance updates · Commure
Claude Sonnet 5.573.3RCM-Bench-1 results: Claude Sonnet 5.5, Insurance updates · Commure
DeepSeek V4.1 Flash73.3RCM-Bench-1 results: DeepSeek V4.1 Flash, Insurance updates · Commure
GPT-6 Luna73.3RCM-Bench-1 results: GPT-6 Luna, Insurance updates · Commure
Grok 4.773.3RCM-Bench-1 results: Grok 4.7, Insurance updates · Commure
Haiku 4.560.0RCM-Bench-1 results: Claude Haiku 4.5, Insurance updates · Commure
GPT-5.6 Luna60.0RCM-Bench-1 results: GPT-5.6 Luna, Insurance updates · Commure
GPT-6 Sol46.7RCM-Bench-1 results: GPT-6 Sol, Insurance updates · Commure
Loading Atlas data…