- Comparisons
Which large language model for which clinical task
A task-first guide to choosing between language models, and what to check before you paste anything.
PDF · 10 pages · Updated 30 September 2026
- Who it is for
- Clinicians who use language models at work and want to choose by task, and to know what they may paste.
- Format
- Length
- 10 pages
- Updated
- 30 September 2026
- Topic
- Comparisons
Three findings inside
5%
of 519 studies of health-care language models used real patient-care data.
Systematic review, JAMA
2 of 3
leaders of the 2025 MedHELM paper no longer appear on its 14 May 2026 leaderboard.
MedHELM leaderboard, 14 May 2026
3 of 3
evaluations were won by general-purpose models over two specialized clinical tools in one blinded benchmark.
Blinded benchmark, Nature Medicine, 2026
What is inside
- How to use the guide
- A task-first decision tableLooking things up, drafting notes, summarising, messages, coding.
- Open-weight or closed: a trade-off, not a ranking
- Data handling: four patterns
- What a benchmark score does and does not show
- Before you paste anything
Get it
A task-first guide to choosing between language models, and what to check before you paste anything.
PDF · 10 pages. The download starts on this page and the file is also sent to your inbox.
Read the guides behind it
Questions
Is this clinical or regulatory advice?
No. It appraises published evidence and sources. Confirm anything that affects a patient or a purchase with your own governance team and counsel.
Do I have to join the briefing to download?
No. The download does not depend on the box. The box is only for the AIMOCS briefing, and nothing is sent until you confirm from the email we send you.
What do you do with my email address?
We use it to send you this file. The privacy notice says what else we do, and how to ask us to delete it.
Physicians and teams working on AI in healthcare
Meet them in the network.
Which large language model for which clinical task
Get it