Meta: healthcare benchmark results
4 models · 5 results · updated August 16, 2026
Everything the index holds for Meta, gathered in one place: Muse Spark, Muse Spark (2026-04-08), Muse Spark 1.2, Muse Spark 1.1. The lab holds first place on HealthBench Hard. Scores sit on each benchmark's own scale and never compare across rows from different boards.
Every result
| model | benchmark | score | position |
|---|---|---|---|
| Muse Spark | HealthBench Hard | 0.428 | 1 of 9 |
| Muse Spark | MedXpertQA (MM) | 78.4% | 3 of 8 |
| Muse Spark (2026-04-08) | MedHELM | 0.621 | 3 of 11 |
| Muse Spark 1.2 | MedScribe (Vals AI) | ~90 | 2 of 84 |
| Muse Spark 1.1 | MedScribe (Vals AI) | ~90 | 3 of 84 |
Which healthcare benchmarks does Meta appear on?
As of August 16, 2026, Meta models hold 5 current results across 4 tracked benchmarks, through Muse Spark, Muse Spark (2026-04-08), Muse Spark 1.2, Muse Spark 1.1.
Where does Meta lead?
Meta models hold first place on HealthBench Hard (Muse Spark, 0.428).
Other labs with pages: Anthropic, OpenAI, Google, Moonshot AI, Alibaba, xAI. The full field is on the index.