Summary
These are the three names most people mean when they say ambient scribe. All three are good products, and they are good at different things, so the honest comparison is about fit rather than a winner.
One thing separates them on evidence. Nabla holds the only randomised trial win in the category. One separates them on transparency. None of the three publishes a straightforward price. And one thing unites them. All three stop at the note.
Here is each on its own terms, then the part the three-way comparison misses.
Comparing scribes is comparing a sixth of the job. See the whole clinic on one record in thirty minutes.
Book a demoAbridge, the enterprise choice
Abridge is the enterprise leader. Best in KLAS for ambient AI in 2025 and 2026, more than 300 health systems, and the deepest Epic integration in the category. If you are a health system on Epic with a procurement function, it is the strongest purchase on this list.
The cost of that is access. No published price, a reported figure near 2,500 dollars a clinician a year, and a three to six month procurement. For a small practice the best enterprise product is one it cannot really buy.
Suki, the voice assistant
Suki is as much a voice assistant as a scribe, with strong dictation and command features layered on the ambient note. Clinicians who want to talk to the chart, not just have it written, often prefer it.
It is reported in the 299 to 399 dollars a month range with setup fees and a year up front, none of it published. Confirm any number directly, because the public figures are estimates.
Nabla, the evidence and privacy pick
Nabla carries the category's only randomised trial, which cut note-writing time 9.5 percent, about 41 seconds an encounter, across roughly 72,000 encounters, and it is built to not store the encounter. If evidence or data residency is what decides it for you, Nabla is the rational choice.
The trial also found neither arm reduced after-hours record time, which is a useful reality check on what a scribe alone changes.
What the evidence says about all three
Only Nabla has a randomised trial, so Abridge and Suki accuracy claims, like ours, are internal measurements rather than trial results. Simulated-encounter research puts average scribe hallucination near 7 percent across the category, worst on the physical exam.
The practical consequence is the same for all three. The clinician reads and signs every note. No product on this list removes that.
The half all three leave
Abridge, Suki and Nabla all cover the consultation and stop. None answers your phone, books the patient, drafts the referral letter, codes the visit or backfills a cancellation. In a health system a different team owns that. In a private clinic it is the same person, after hours.
That is where WA\ is a different shape. The note, the letters with citations, the coding and the front desk on one record, with a published price and a fourteen day trial. We are not claiming to beat Abridge on Epic depth or Nabla on trial evidence. We are claiming to cover the five sixths a scribe leaves behind.
The short answer
Enterprise on Epic, Abridge. Want a voice assistant, Suki. Want the evidence and privacy pick, Nabla. Running a private clinic where the note is only part of the evening, look past the scribe question entirely to the system that covers the whole day, which is WA\.
Frequently asked questions
Which is best, Abridge, Suki or Nabla?
They win on different criteria, so the best one is the one that matches you. Abridge is the enterprise Epic choice with Best in KLAS and 300 plus health systems. Suki is the voice-assistant pick with strong dictation. Nabla holds the category's only randomised trial and is built not to store the encounter, so it is the evidence and privacy choice. None publishes a simple price, and all three stop at the note rather than covering the letters, coding and front desk.
Which ambient scribe has the best evidence?
Nabla. It is the only one of the three with a randomised controlled trial, which cut note-writing time by 9.5 percent, about 41 seconds per encounter, across roughly 72,000 encounters, though neither arm reduced after-hours record time. Abridge and Suki publish internal accuracy figures rather than trial results, as most vendors do, including us. Whatever you choose, average scribe hallucination near 7 percent means the clinician signs every note.
Do any of them do more than the note?
Not really, and that is the key point for a private clinic. Abridge, Suki and Nabla all cover the consultation and stop, leaving the phone, the booking, the referral letter, the coding and the cancellation to someone else. In a private clinic that someone is usually the same person, after hours. WA\ is built to cover that whole tail on one record, with a published price, which is a different product from a scribe.
Live this week, not next quarter.
Thirty minutes and you will see the consult, the front desk and the patient record running as one system.