MedScribe (Vals AI): current results
Vals AI (dataset with Protege) · 100 rubric-scored SOAP-note cases · index updated August 16, 2026
Claude Opus 5 holds the top current result on MedScribe (Vals AI), 90.99% as of 2026-08, per Vals AI MedScribe leaderboard. Clinical documentation support: quality of SOAP notes generated from clinical visits, scored against rubrics for documentation quality and compliance.
Current results
- 1
Claude Opus 590.99% - 2
Muse Spark 1.2~90 - 3Muse Spark 1.1~90
- 4Claude Fable 5~90
- 5
GPT 5.188.09% - 6Claude Opus 4.686.74%
Result detail
| # | model | score | as of | |
|---|---|---|---|---|
| 1 | | Claude Opus 5 Anthropic | 90.99% | 2026-08 |
| 2 | | Muse Spark 1.2 Meta rank 2; exact value not displayed by the source | ~90 | 2026-08 |
| 3 | | Muse Spark 1.1 Meta rank 3; exact value not displayed by the source | ~90 | 2026-08 |
| 4 | | Claude Fable 5 Anthropic rank 4; exact value not displayed by the source | ~90 | 2026-08 |
| 5 | | GPT 5.1 OpenAI at-release leader (Feb 2026 blog) | 88.09% | 2026-02 |
| 6 | | Claude Opus 4.6 Anthropic at-release (Feb 2026 blog) | 86.74% | 2026-02 |
Scores appear exactly as Vals AI MedScribe leaderboard publishes them (independently run). Vals AI self-runs; 84 models, last updated August 15, 2026. Top scores cluster near 90, so the leaders sit close to the ceiling, and exact decimals below first place are not always displayed.
About the benchmark
| publisher | Vals AI (dataset with Protege) |
|---|---|
| category | documentation and coding benchmarks |
| released | 2026-02 |
| size | 100 rubric-scored SOAP-note cases |
| scale | percentage accuracy 0-100, higher better |
| result basis | independently run |
| source | Vals AI MedScribe leaderboard |
| last frontier result | 2026-08 |
What is MedScribe (Vals AI)?
MedScribe (Vals AI) is a documentation and coding benchmark from Vals AI, released 2026-02: 100 rubric-scored SOAP-note cases, scored on a percentage accuracy 0-100 scale. Clinical documentation support: quality of SOAP notes generated from clinical visits, scored against rubrics for documentation quality and compliance.
Which model leads MedScribe (Vals AI)?
Claude Opus 5 (Anthropic) holds the top current result on MedScribe (Vals AI) at 90.99%, per Vals AI MedScribe leaderboard, as of 2026-08.
Where do the MedScribe (Vals AI) numbers come from?
From Vals AI MedScribe leaderboard (independently run). Vals AI self-runs; 84 models, last updated August 15, 2026. Top scores cluster near 90, so the leaders sit close to the ceiling, and exact decimals below first place are not always displayed.
The rest of the field is on the index, and how sources qualify is on the methodology page. Model names in the table link to cross-benchmark pages.