Kimi K2.5: healthcare benchmark results
Moonshot AI · 5 boards · snapshot reviewed September 28, 2026
The index currently holds 5 results for Kimi K2.5: 39.32% on MedCode (Vals AI) (64 of 102 indexed rows), 76.44% on MedScribe (Vals AI) (71 of 104 indexed rows), 65.3 on MedXpertQA (MM) (14 of 22 indexed rows), 0.62 on EHR-Complex [Kimi-K2.5] (3 of 18 indexed rows), 15.6% on HealthAdminBench (3 of 7 indexed rows). Scores below sit on different scales and come from different graders, so read each against its own benchmark, never against the others.
Results by benchmark
| benchmark | score | index position | as of |
|---|---|---|---|
| MedCode (Vals AI) model ID kimi/kimi-k2.5-thinking; top_p=0.95; max_output_tokens=30000; standard error 2.119 pp; $0.017275/test; source snapshot 2026-09-26; run date not published | 39.32% | 64 of 102 | 2026-09-26 |
| MedScribe (Vals AI) model ID kimi/kimi-k2.5-thinking; top_p=0.95; max_output_tokens=30000; standard error 1.986 pp; $0.024890/test; source snapshot 2026-09-26; run date not published | 76.44% | 71 of 104 | 2026-09-26 |
| MedXpertQA (MM) Qwen-run comparison in the Qwen3.5-397B-A17B model card (K2.5-1T-A32B column) | 65.3 | 14 of 22 | not reported |
| EHR-Complex Kimi-K2.5 headline 12-model evaluation, top open-weight via paper, arxiv.org | 0.62 | 3 of 18 | 2026-06 |
| HealthAdminBench screenshot-only, task description + portal guidance via paper, arxiv.org | 15.6% | 3 of 7 | 2026-04 |
Position follows the rows held in this index and is not a controlled comparison. Sources may use different graders or task subsets. Positions include separate configurations. A source board may contain additional rows; its size is listed separately when it differs. Config caveats, where a source noted any, are on each benchmark's page, and the document behind each score is cited on the sources page. Sources also list this model as "Kimi-K2.5".
Which healthcare benchmarks is Kimi K2.5 scored on?
As of September 28, 2026, Kimi K2.5 has indexed results on 5 tracked benchmarks: MedCode (Vals AI), MedScribe (Vals AI), MedXpertQA (MM), EHR-Complex, HealthAdminBench.
Which results are indexed for Kimi K2.5?
Kimi K2.5 stands at 39.32% on MedCode (Vals AI) (64 of 102 indexed rows), 76.44% on MedScribe (Vals AI) (71 of 104 indexed rows), 65.3 on MedXpertQA (MM) (14 of 22 indexed rows), 0.62 on EHR-Complex [Kimi-K2.5] (3 of 18 indexed rows), 15.6% on HealthAdminBench (3 of 7 indexed rows).
The benchmarks themselves are described on their pages, linked in the table above, and the whole field is on the index.