A harder subset of OpenAI's HealthBench for evaluating open-ended medical and health reasoning with rubric-based grading.
BenchLM mirrors the published score view for HealthBench Hard. Muse Spark leads the public snapshot at 42.8% , followed by GPT-5.4 (40.1%) and Gemini 3.1 Pro (20.6%). BenchLM does not use these results to rank models overall.
Muse Spark
Meta
GPT-5.4
OpenAI
Gemini 3.1 Pro
The published HealthBench Hard snapshot is tightly clustered at the top: Muse Spark sits at 42.8%, while the third row is only 22.2 points behind. The broader top-10 spread is 28.0 points, so the benchmark still separates strong models even when the leaders cluster.
5 models have been evaluated on HealthBench Hard. The benchmark falls in the Knowledge category. This category carries a 12% weight in BenchLM.ai's overall scoring system. HealthBench Hard is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
Year
2026
Tasks
1,000 health prompts
Format
Open-ended health evaluation
Difficulty
Advanced health reasoning
Meta describes HealthBench Hard as a 1,000-prompt subset of OpenAI's HealthBench, graded with the same simple-evals implementation and a GPT-4.1-based judge. BenchLM treats it as a display-only health benchmark reference.
Version
HealthBench Hard 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
A harder subset of OpenAI's HealthBench for evaluating open-ended medical and health reasoning with rubric-based grading.
Muse Spark by Meta currently leads with a score of 42.8% on HealthBench Hard.
5 AI models have been evaluated on HealthBench Hard on BenchLM.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.