Heidi
Also known as: Heidi Health, Heidi AI
AI care partner for clinicians: an ambient scribe with writeback to Epic, Oracle Cerner and athenahealth, agents and scheduled Routines that do the admin around each visit and wait for review before anything touches a chart or a patient, Heidi Evidence for cited answers and Heidi Comms for patient calls.
Heidi, founded in Melbourne, builds an AI care partner for clinicians in more than 190 countries, and its ambient scribe is used about 3 million times a week. The scribe captures consultations on phone, desktop (Heidi Desktop for macOS and Windows, which also dictates anywhere on the screen and captures telehealth audio) or the Heidi Remote wearable and drafts notes from standard and advanced templates.
Heidi II, released in September 2026, adds agents that do the work around the visit. A clinician asks in plain language, and Heidi works the task through across integrated systems and legacy portals, from preparing tomorrow's ward round from the charts overnight to chasing pathology, filling a referral form and drafting a handover letter.
A saved task becomes a Routine that runs on a schedule for one clinician or a whole practice, and more than twenty ready made Routines, built by clinicians, cover morning prep, visit paperwork, results follow up, insurer forms, eligibility checks through Availity, compliance checks and inbox triage. Memory learns each clinician's preferences from what they say and every correction they make. Heidi starts in Manual mode, and by default anything that touches a chart, reaches a patient or leaves the system waits for review, with Auto mode for tasks the clinician trusts. Heidi will not prescribe, delete a record or submit a claim in any mode. Heidi II is rolling out in English and is not available in the EU and UK.
Heidi Evidence answers clinical questions from the full text of journals including the New England Journal of Medicine, Wiley and Cochrane, alongside the patient's history, with citations. Heidi Comms agents answer patient calls and messages around the clock, handle routine enquiries, bookings, reminders and follow ups by phone and SMS, log every interaction with its transcript and outcome, and transfer the call to reception when a patient asks for a person.
Integrations write to the chart in Epic (structured writeback and clinical task automation across Hyperspace, Hyperdrive and Haiku), Oracle Cerner (MPage components in PowerChart and FirstNet) and athenahealth (schedule sync and push to chart), with Altera, AlayaCare, eClinicalWorks, Veradigm and others, plus Gmail, Outlook, Slack, Teams, Zoom and Meet. A public Heidi API and Widget SDK serve integration partners outside Australia and New Zealand, with regions for AU, US, EU, CA and UK.
Heidi lists SOC 2 Type II, ISO 27001, ISO 42001, TX-RAMP Level 2, HIPAA, GDPR and Cyber Essentials Plus. Plans are Free, Clinician (Standard or Max seats, with a 14 day trial) and Teams by quote. In September 2026 Heidi raised US$340 million led by General Catalyst's Customer Value Fund, Blackbird and Headline, bringing total funding to US$436.6 million.
From our comparison video
Heidi vs Nabla: 7 Questions for Clinicians Choosing an AI Scribe
Our 7 minute video puts Heidi side by side with Nabla. It covers what each one does and how each is priced, then when to choose which.
Recorded in October 2026. The profile here is live and updates as research is verified.
Vendor details
Canonical URL
https://www.heidihealth.com
Category
Healthcare agent
Subcategory
Clinical documentation
Funding status
Independent, headquartered in Melbourne, Australia. Raised US$340 million in September 2026 (Series C and growth investment) led by General Catalyst's Customer Value Fund, Blackbird and Headline, for US$436.6 million raised to date; reported annual recurring revenue of US$50 million as of April 2026.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Writes to the chart in Epic (structured writeback and clinical task automation across Hyperspace, Hyperdrive and Haiku), Oracle Cerner (MPage components in PowerChart and FirstNet) and athenahealth (schedule sync and push to chart); also Altera, AlayaCare, Best Practice, Medical Director, eClinicalWorks, Veradigm, CareStack and others, Availity for eligibility, and Gmail, Google Calendar and Drive, Outlook, OneDrive and SharePoint. Full EHR integration access sits on the Practice plan, lite access on Clinician. Heidi API and Widget SDK for partners, closed to new integrations in Australia and New Zealand.
In practice
Your clinicians lose the end of every day to referral letters, results chasing and forms. Heidi's agents work through that admin across the chart and legacy portals, and nothing reaches a chart or a patient until the clinician approves it.
You want evidence at the point of care, not in a separate tab. Heidi Evidence answers clinical questions from the full text of journals such as NEJM, Wiley and Cochrane, alongside the patient's history, with citations.
A resident wants to trial an AI scribe without a budget line. Heidi's free plan includes unlimited transcription with standard templates, enough to run real clinic days before paying.
Sources & related URLs
Agentic Index coverage score
10.5 / 14 capabilities · 75%
| Integrations & Tool Calling | Full |
|---|---|
|
Heidi writes back to the chart through named EHR integrations. In Epic it offers structured writeback and clinical task automation across Hyperspace, Hyperdrive and Haiku, in Oracle Cerner it writes back to MPage components in PowerChart and FirstNet, and with athenahealth it syncs schedules and pushes to chart. Altera, AlayaCare, eClinicalWorks, Veradigm and others are also named, along with Availity eligibility checks. Agents also reach Gmail and Outlook, Slack and Teams channels, and Zoom, Meet and Teams calls. Full EHR access sits on the Teams plan. Sourceheidihealth.com/en-us/integrationsread 2026-10-23 |
|
| Workflow Orchestration | Full |
|
Agents take a task asked in plain language and work it through to the end across integrated systems and legacy portals, such as preparing tomorrow's ward round from the charts overnight, chasing outstanding pathology, filling a referral form from the letter that came with it and drafting a handover letter. A saved task becomes a Routine that runs the same way every time, on a schedule, for one clinician or a whole practice, and anything that could not be finished comes back named with a reason. Agents and Routines draw on a monthly usage allowance that comes with each paid seat. Sourceheidihealth.com/en-us/blog/heidi-ii-is-hereread 2026-10-23 |
|
| Knowledge Grounding & RAG | Partial |
|
Heidi Evidence answers clinical questions from the full text of articles in the New England Journal of Medicine, Wiley, Cochrane and other journals, alongside the patient's history, with citations a journal subscriber can follow. Patient context comes from the EHR, an integration or a pasted list, and agents cite every line of a prep note back to the record. Evidence draws on literature Heidi curates. The plans list a personal evidence library and, on Teams, an org wide library and team guidelines, but Heidi does not describe what goes into them. Sourceheidihealth.com/en-us/blog/heidi-ii-is-hereread 2026-10-23 |
|
| Human Oversight & Guardrails | Full |
|
Heidi starts in Manual mode, where the clinician watches each run and approves as it goes, and by default anything that touches a chart, reaches a patient or leaves the system waits for review. A run can be stopped at any point, and Auto mode lets a task the clinician trusts run on its own. In every mode Heidi will not prescribe, delete a record or submit a claim. Nothing goes to the patient until the clinician has read the recap and approved it, and the Comms agent hands the call to reception as soon as a patient asks for a person. Sourceheidihealth.com/en-us/blog/heidi-ii-is-hereread 2026-10-23 |
|
| Security, Identity & Governance | Full |
|
Certifications listed include SOC 2 Type II, ISO 27001, ISO 42001, TX-RAMP Level 2, HIPAA, GDPR, Cyber Essentials Plus and UK Class I medical device status, and Heidi II ships under the same independently audited certifications. SSO and enterprise governance come with the Teams plan, and the reports themselves are requested through Heidi's trust center. Sourceheidihealth.comread 2026-10-23 |
|
| Observability & Auditability | Full |
|
Every Comms interaction is logged with its transcript and outcome, and every call keeps a full transcript and recording for review. A Routine skips nothing silently, and anything Heidi could not finish comes back named with a reason. Heidi shows no run log or audit export for what an agent did. Sourcesupport.heidihealth.com/en/articles/13396153-heidi-commsread 2026-10-23 |
|
| Memory & State Persistence | Partial |
|
Memory learns each clinician's unwritten rules, clinical standards and formatting preferences, from what they tell Heidi directly and from every correction they make, and it draws on the seat's usage allowance. A word library, pinned templates and note insights, which group the corrections a clinician keeps making and suggest word library entries, auto edit rules and template changes to approve, also carry across sessions. Heidi states no lifetime for what Memory keeps and no way to view or delete it. Sourceheidihealth.com/en-us/blog/heidi-ii-is-hereread 2026-10-23 |
|
| Deployment & Data Residency | Full |
|
Regions are mapped for AU, US, EU, CA and UK, and custom hosting is offered on Enterprise. Heidi does not say how a customer's region gets assigned. Heidi II, the agent release, is rolling out globally in English and is not available in the EU and UK. Sourceheidihealth.com/developers/additional-resources/region-mappingread 2026-10-23 |
|
| Prebuilt Agents, Templates & Packs | Full |
|
Ready made Routines, built by clinicians who run them every day, can be switched on as they are. Heidi names more than twenty, among them morning prep, visit paperwork, follow up digest, patient recap, ward round prep, shift handover, results follow up, form filler, insurer forms, an eligibility check through Availity, a compliance check and inbox triage. Standard and advanced note templates sit beside them, and Scribe, Evidence and Comms remain separate products. Sourceheidihealth.com/en-usread 2026-10-23 |
|
| Triggers & Channel Coverage | Full |
|
Routines run on a schedule the clinician or practice sets, and Heidi Comms agents answer inbound patient calls and messages around the clock and run outbound reminders and follow ups by phone and SMS. Agents can also read the morning's appointments and open a session for each patient, and athenahealth schedules sync automatically. Sourcesupport.heidihealth.com/en/articles/13396153-heidi-commsread 2026-10-23 |
|
| Model Flexibility & Routing | Not documented |
|
No model provider is named, and the customer has no model choice or routing. Sourceheidihealth.com/developersread 2026-10-23 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
The Heidi API, with public developer documentation, covers sessions with context and coding, transcription, consult and sectional notes, documents, templates, patient profiles and Ask Heidi. A Heidi Widget comes with an SDK, plus a beta Usage API. The API and Widget are closed to new integrations in Australia and New Zealand. Sourceheidihealth.com/developersread 2026-10-23 |
|
| Testing, Debugging & Optimization | Not documented |
|
Customers get no evaluation or testing surface for the scribe, the agents or Routines. The safety research Heidi publishes, including a 2026 study of 385 simulated visits and an automated judge checked against clinicians' own ratings, is its own testing. Sourceheidihealth.comread 2026-10-23 |
|
| Browser & Computer Use | Partial |
|
Computer use is one of the Agents features on every plan, drawing on the seat's monthly usage allowance, and Heidi says its agents work tasks through across integrated systems or legacy portals. It does not describe how an agent operates a portal, which systems it can drive or how a session is recorded. Heidi Desktop also lets the clinician dictate anywhere on the screen and captures telehealth audio straight from the call. Sourceheidihealth.com/en-us/pricingread 2026-10-23 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Heidi II began rolling out in English and French, turning the ambient scribe into an agent that takes a plain language request and carries it through, such as preparing the day's charts, chasing pathology results and filling referral forms. Saved tasks become Routines that run on a schedule, on an event or across a patient list. Clinicians choose between approving every step and letting routine work run, and prescribing, claims, deleting records and diagnosis always wait for sign off.
Bears on: Human approval / guardrails
View sourceHeidi Health launched its ambient AI medical scribe platform in Hong Kong. This regional expansion introduces its automated clinical documentation tools to a new market.
Bears on: Deployment / data residency
View sourcePricing
Free plan; Clinician by Standard or Max seat (the $150 per user per month on record dates from July and is unconfirmed); Teams by quote
seats (per user per month)
Included quota
Free: one seat, unlimited transcription, standard templates, task management and dictation, with limited Ask, advanced templates, coding, Evidence and agents on a weekly allowance. Clinician: a Standard or Max seat with a monthly usage allowance for Agents, Routines, computer use, Memory and connectors (Max carries five times the usage), custom templates and evidence inside the visit. Teams: custom seat types with shared monthly capacity, EHR integration, team template sharing, SSO and custom terms and hosting.
What is public
Heidi publishes a pricing page with the plan table (Free, Clinician with Standard or Max seats, and Teams) and prices in local currency. The Clinician rate last recorded, $150 per user per month billed annually, dates from July, before the Standard and Max seats.
Billing mechanics
Self serve per user subscriptions with monthly or yearly billing on Clinician, as a Standard or a Max seat. Teams is quoted, with seat types mixed on one invoice and shared monthly capacity.
Cost watchouts
Daily agent use can outrun a Standard seat's allowance, which is what the Max seat is for, and EHR integration and SSO sit on the quoted Teams plan.
Variable cost rationale
Flat per seat pricing with unlimited scribing and Evidence. Agent work draws on a usage allowance per seat, and heavier daily use moves a clinician from a Standard to a Max seat rather than into metered overage.
Additional watchouts
Plan names changed again with Heidi II in September 2026 (Free, Clinician and Teams, with Standard and Max seats), and prices vary by currency, so confirm the local rate on the pricing page.
Overage / add-ons
Agents, Routines, computer use, Memory and connectors draw on a usage allowance, weekly on Free and monthly on paid seats, and Max seats carry five times the Standard allowance. Scribing and Evidence questions never use it. Teams shares monthly capacity with options to expand.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free plan; 14 day free trial on Clinician
Lowest paid plan
Clinician, as a Standard seat; the last recorded rate, $150 per user per month billed annually, is from July and unconfirmed
Commercial notes
Independent, Melbourne headquartered; US$340 million raised in September 2026 led by General Catalyst's Customer Value Fund, Blackbird and Headline, US$436.6 million to date.
Key ambiguities
Current Clinician rates for the Standard and Max seats are not confirmed. The pricing page shows local currency prices that did not render, and the $150 figure on record predates the Heidi II seat structure.
Related vendors
- Abridge — AI platform for health systems that turns clinical conversations…
- Adonis — AI orchestration platform for revenue cycle management whose…
- AKASA — Generative AI for the hospital mid revenue cycle: Coding and CDI…
- Alaffia Health — Agentic AI for health plan claims operations: utilization…
- AlethianAI — Connected AI agents for medical practices share one patient record
- Ambience Healthcare — AI platform for health systems pairing an ambient scribe with…
Alternatives to Heidi
The closest documented capability profiles to Heidi among healthcare agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Scribeberry10.5 / 14Adds documented Model Flexibility & Routing
- Prosper AI10.0 / 14Adds documented Testing, Debugging & Optimization
- Microsoft Dragon Copilot9.0 / 14Adds documented Testing, Debugging & Optimization
- SPRY9.0 / 14Adds documented Testing, Debugging & Optimization
- SuperDial9.0 / 14Adds documented Testing, Debugging & Optimization
- XCaliber Health9.0 / 14Fuller documented coverage on Knowledge Grounding & RAG
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded