MedScribe (Vals AI) leaderboard
- Claude Opus 5.5 leads
- 91.4%
- models measured
- 85
- lead over #2
- +0.1 pts
- median
- 83.0%
Leader on MedScribe (Vals AI) · read October 22
On October 22, 2026, 85 of our catalog's models had a MedScribe (Vals AI) result, and Anthropic Claude Opus 5.5 led with 91.4%, ahead of Anthropic Claude Fable 5.1 at 91.3%.
Writing clinical SOAP notes, graded section by section. The score as Vals AI publishes it.
Role in the index
MedScribe (Vals AI) is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Latency, as printed | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic | 91.4% | 7m 22s | 88.8 | $20.00 | |
| 2 | Claude Fable 5.1Anthropic | 91.3% | 3m 08s | 84.2 | $50.00 | |
| 3 | Claude Sonnet 5.5Anthropic | 91.1% | 5m 07s | 87.3 | $10.00 | |
| 4 | Claude Opus 5Anthropic | 91.0% | 1m 17s | 83.0 | $25.00 | |
| 5 | Muse Spark 1.2Meta | 90.1% | 1m 01s | 55.7 | $4.25 | |
| 6 | Claude Haiku 5.5Anthropic | 89.5% | 5m 23s | 70.7 | $0.50 | |
| 7 | Grok 4.7xAI | 89.4% | 2m 33s | 66.7 | $6.00 | |
| 8 | GLM 5.3 FlashZ.ai | 88.9% | 1m 27s | 57.7 | $0.50 | |
| 9 | Muse Spark 1.1Meta | 88.9% | 1m 03s | 70.3 | $4.25 | |
| 10 | GLM 5.3Z.ai | 88.8% | 2m 02s | 70.6 | $4.40 | |
| 11 | Claude Fable 5Anthropic | 88.5% | 1m 59s | 79.3 | $50.00 | |
| 12 | MiMo-V2.6-ProXiaomi | 88.3% | 2m 53s | 65.4 | $0.87 | |
| 13 | GPT-5.1OpenAI | 88.1% | 1m 18s | 40.4 | $10.00 | |
| 14 | Kimi K3Moonshot AI | 88.0% | 38 s | 62.3 | $15.00 | |
| 15 | GPT-6 AstraOpenAI | 87.9% | 2m 34s | 85.1 | $50.00 | |
| 16 | Gemini 4 ArgonGoogle | 87.4% | 1m 27s | 81.0 | $10.00 | |
| 17 | MiniMax M3MiniMax | 87.3% | 2m 04s | 39.2 | $1.20 | |
| 18 | Grok 4.5xAI | 86.9% | 11m 45s | 58.6 | $6.00 | |
| 19 | GPT-5.5OpenAI | 86.9% | 2m 13s | 71.3 | $30.00 | |
| 20 | Claude Opus 4.6Anthropic | 86.7% | 54 s | 57.2 | $25.00 | |
| 21 | Grok 4.6xAI | 86.5% | 1m 11s | 65.7 | $6.00 | |
| 22 | GPT-6.1 SolOpenAI | 86.5% | 2m 50s | 82.4 | $10.00 | |
| 23 | Claude Opus 4.8Anthropic | 85.8% | 1m 22s | 65.0 | $25.00 | |
| 24 | DeepSeek V4.1 FlashDeepSeek | 85.5% | 54 s | 62.5 | $0.84 | |
| 25 | T InklingThinking Machines | 85.4% | 4m 05s | 31.5 | $4.05 | |
| 26 | Claude Opus 4.5Anthropic | 85.3% | 1m 12s | 44.0 | $25.00 | |
| 27 | MiMo-V2.6-FlashXiaomi | 85.3% | 1m 14s | 54.7 | $0.28 | |
| 28 | Claude Haiku 4.5Anthropic | 85.2% | 1m 06s | 35.6 | $5.00 | |
| 29 | GPT-5.6 SolOpenAI | 85.2% | 1m 34s | 73.5 | $20.00 | |
| 30 | Qwen3.8 Max (0902)Qwen | 85.0% | 4m 11s | 62.4 | $6.00 | |
| 31 | Claude Sonnet 4.5Anthropic | 84.5% | 44 s | 39.5 | $15.00 | |
| 32 | Gemini 3.8 FlashGoogle | 84.5% | 21 s | 61.6 | $3.75 | |
| 33 | GPT-5.2OpenAI | 84.4% | 2m 06s | 55.9 | $14.00 | |
| 34 | GPT-5.6 LunaOpenAI | 84.4% | 1m 57s | 58.0 | $1.20 | |
| 35 | T Inkling SmallThinking Machines | 84.1% | 3m 50s | 27.3 | $1.20 | |
| 36 | Gemini 3.7 FlashGoogle | 83.9% | 17 s | 58.0 | $3.75 | |
| 37 | Qwen3.8 27BQwen | 83.8% | 1m 30s | 74.2 | $2.55 | |
| 38 | MiMo-V2.5-ProXiaomi | 83.7% | 1m 31s | 31.6 | $0.9152 | |
| 39 | GPT-6 LunaOpenAI | 83.7% | 2m 50s | 57.3 | $0.50 | |
| 40 | GPT-5OpenAI | 83.7% | 3m 05s | 38.8 | $10.00 | |
| 41 | Hy4 previewTencent | 83.6% | 5m 50s | 44.3 | $2.501 | |
| 42 | GLM 5.2Z.ai | 83.5% | 2m 18s | 61.1 | $4.40 | |
| 43 | Gemini 2.5 FlashGoogle | 83.0% | 23 s | Not benchmarked | $2.50 | |
| 44 | Claude Opus 4.7Anthropic | 83.0% | 1m 08s | 67.5 | $25.00 | |
| 45 | GPT-5.6 TerraOpenAI | 82.9% | 36 s | 63.9 | $12.00 | |
| 46 | GPT-6 SolOpenAI | 82.0% | 1m 21s | 76.0 | $10.00 | |
| 47 | Ling 3.0 FlashinclusionAI | 80.9% | 12 s | 29.6 | $0.1215 | |
| 48 | MiniMax M2.1MiniMax | 80.8% | 53 s | Not benchmarked | $1.20 | |
| 49 | GPT-5 MiniOpenAI | 80.6% | 4m 36s | 34.2 | $2.00 | |
| 50 | Mistral Large 4Mistral AI | 80.4% | 2m 05s | 56.4 | $4.18 | |
| 51 | DeepSeek V4 Flash 0731DeepSeek | 80.4% | 1m 18s | 51.8 | $0.396 | |
| 52 | DeepSeek V4 Pro 0813DeepSeek | 80.2% | 2m 36s | 56.1 | $1.98 | |
| 53 | MiniMax M2.7MiniMax | 79.9% | 27 s | 36.2 | $1.20 | |
| 54 | Gemini 3.6 FlashGoogle | 79.7% | 34 s | 62.5 | $3.75 | |
| 55 | Qwen3.7 MaxQwen | 79.4% | 1m 48s | 54.1 | $4.425 | |
| 56 | Kimi K2.6Moonshot AI | 78.2% | 8m 14s | 52.1 | $4.00 | |
| 57 | GPT-5.4OpenAI | 77.5% | 4m 43s | 64.4 | $15.00 | |
| 58 | GPT-5.4 NanoOpenAI | 77.1% | 21 s | 44.8 | $1.25 | |
| 59 | Qwen3.6 PlusQwen | 77.0% | 2m 54s | 44.1 | $1.95 | |
| 60 | o3OpenAI | 76.7% | 46 s | Not benchmarked | $8.00 | |
| 61 | Gemini 3.5 FlashGoogle | 76.6% | 57 s | 39.8 | $9.00 | |
| 62 | Kimi K2.5Moonshot AI | 76.4% | 2m 27s | 38.6 | $2.85 | |
| 63 | Gemini 3.1 Pro PreviewGoogle | 76.1% | 1m 09s | 32.3 | $12.00 | |
| 64 | Claude Sonnet 5Anthropic | 76.0% | 4m 12s | 57.7 | $10.00 | |
| 65 | Gemini 2.5 Flash LiteGoogle | 75.8% | 4.5 s | Not benchmarked | $0.40 | |
| 66 | DeepSeek V4 Pro 0423DeepSeek | 75.1% | 5m 46s | 38.7 | $2.976 | |
| 67 | Grok 4.3xAI | 74.4% | 1m 40s | 37.4 | $2.50 | |
| 68 | Claude Opus 4.1Anthropic | 73.9% | 57 s | Not benchmarked | $75.00 | |
| 69 | Gemini 2.5 ProGoogle | 73.5% | 36 s | 19.1 | $10.00 | |
| 70 | GPT-5 NanoOpenAI | 72.9% | 1m 53s | Not benchmarked | $0.40 | |
| 71 | GLM 5.1Z.ai | 72.3% | 1m 36s | 49.2 | $4.40 | |
| 72 | MiMo-V2.5Xiaomi | 72.2% | 20 s | 50.0 | $0.336 | |
| 73 | Gemini 3.5 Flash LiteGoogle | 70.9% | 19 s | 49.0 | $2.50 | |
| 74 | Qwen3.5-FlashQwen | 70.6% | 1m 20s | 41.1 | $0.26 | |
| 75 | Gemini 3 Flash PreviewGoogle | 69.9% | 23 s | 43.7 | $3.00 | |
| 76 | o4 MiniOpenAI | 69.1% | 1m 22s | Not benchmarked | $4.40 | |
| 77 | GLM 4.7Z.ai | 68.6% | 2m 47s | 55.1 | $2.20 | |
| 78 | Mistral Medium 3.5Mistral AI | 67.7% | 1m 44s | 25.5 | $7.50 | |
| 79 | Gemini 3.1 Flash Lite PreviewGoogle | 63.9% | 17 s | 22.7 | $1.50 | |
| 80 | Grok 4.20xAI | 63.4% | 19 s | 33.3 | $2.50 | |
| 81 | Command ACohere | 55.7% | 3m 23s | Not benchmarked | $10.00 | |
| 82 | Mercury 2.5Inception | 55.1% | 10.0 s | 1.6 | $0.15 | |
| 83 | Llama 4 MaverickMeta Llama | 54.2% | 25 s | Not benchmarked | $1.00 | |
| 84 | Llama 4 ScoutMeta Llama | 50.6% | 11 s | Not benchmarked | $0.50 | |
| 85 | Nemotron 3.5 LightningNVIDIA | 4.3% | 1m 46s | 22.6 | $0.147 |
Only models in our catalog are listed, and each appears once with its best published run. 137 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.