Freed and Nabla are not clean substitutes on the evidence available. Freed’s public materials describe a clinician-facing documentation product with recording, note generation, template learning, AI chat, and browser-based EHR write-back options, including Chrome-extension note transfer for named systems and newer direct connection flows for at least Athenahealth and eClinicalWorks.[1][2][3][6][8] Nabla’s public materials split between an EHR-facing clinical AI layer and a developer-oriented Core API that can transcribe encounters, generate notes, edit notes from instructions, produce dictation output, and extract FHIR-compliant normalized data.[9][11][12] For a buyer, the main decision point is less “which scribe is better” than “are we buying end-user software for clinicians, or an embedded platform and API strategy?” The public pages do not settle several practical questions, including implementation effort, workflow fit for a given specialty, and like-for-like feature parity across actual deployments.[9][11]
The strongest documented difference is service model, not note-generation basics
Freed’s documentation reads like end-user software for clinicians and practices. Its public pages describe starting a visit, recording through the device microphone, generating a structured note, editing the note, asking chat questions, producing follow-on documentation such as letters or patient instructions, and then sending content into an EHR.[3][4][5] The emphasis is on what a clinician does inside the product and, in several help articles, on how to operate the workflow in Chrome with a specific EHR open behind the Freed window.[1][6][8]
Nabla’s documentation, by contrast, is split across two public faces. One is an EHR integration page describing a clinical AI layer that runs inside an existing EHR on desktop or mobile and writes documentation, dictation, coding outputs, diagnoses, vitals, and patient instructions back into chart fields.[11] The other is the Core API documentation aimed at builders, where the documented units are endpoints for transcription, note generation, note editing, dictation, and normalized data extraction rather than a clinician self-serve workflow.[9][12]
That distinction matters because similar high-level labels can hide different buying motions. A solo or small-group practice evaluating Freed’s public materials can see a direct software workflow and specific browser instructions.[1][7] A health system, EHR vendor, or digital health company reading Nabla’s public materials sees either an embedded EHR layer or an API surface intended to be integrated into another product.[9][11] Publicly, those are different operational categories, even though both touch clinical documentation.
Where Freed is more concrete today: browser workflows and EHR field routing examples
Freed’s public evidence is unusually specific about browser-based note transfer. The help center says EHR push transfers notes into any browser-based EHR through the Chrome extension, while also warning users to review chart accuracy and confirming that the push goes into whichever chart is open.[1] The marketing EHR page lists named systems and says the extension works beside the chart in Chrome, with one-click note transfer and patient detail sync in some described cases.[2]
The Athenahealth and eClinicalWorks guides go further by showing section-level behavior. For Athenahealth, Freed documents where Review of Systems, Objective, Assessment and Plan, and Patient Instructions are placed, with settings that affect how some content is routed.[8] For eClinicalWorks, the guide explains that some content can be concatenated into the HPI field, that exam content follows the active examination category or content matching, and that assessment and plan routing depends on diagnosis matching and user settings.[6]
That level of detail gives buyers concrete workflow questions to test internally, such as whether their note structure matches destination fields and whether staff are comfortable with browser-extension operation and post-push verification.[1][6][8] It does not prove the workflow is preferable, but it does make Freed’s documented chart-write process more inspectable from public materials than a generic claim of EHR compatibility would be.
A practical way to frame an internal evaluation
As a suggestion, a buyer can separate the decision into two tracks before booking demos. Track one is the end-user workflow question: do clinicians need a standalone documentation application with recording, editing, chat-based note support, templates, and browser EHR transfer steps like the ones Freed documents for Athenahealth or eClinicalWorks?[3][5][6][8] Track two is the platform question: do you need an embedded EHR layer or API functions such as note generation, dictation, and FHIR-oriented normalized data that another team will integrate, as Nabla documents?[9][11][12]
That framing helps avoid a false one-to-one comparison. If the organization is comparing deployment models as much as product features, the useful next questions are about ownership, implementation path, and exact chart destinations rather than a generic claim that both are “AI scribes.” The public pages support that distinction clearly, even if they do not answer every operational detail.[1][9][11]
Where Nabla is more concrete today: embeddable architecture and normalized clinical data
Nabla’s most specific public detail is not a click path for an individual clinician but a product architecture for embedding. Its EHR page says the clinical AI layer runs inside the EHR already used by clinicians and that EHR vendors can embed Nabla through a single iframe with new capabilities shipping without re-integration.[11] The same page also lists desktop and mobile use, SMART on FHIR login support, and configurable data-retention language, all in the context of an EHR-integrated deployment.[11]
On the developer side, Nabla’s Core API documentation is explicit about available functions. The introduction page documents medical encounter transcription, clinical note generation from templates such as SOAP, APSO, or WCC, natural-language note editing called Magic Edit, dictation, Dot Phrase suggestions, custom dictionary support, and feedback reporting.[9] The changelog adds dated versioning and newer normalized-data and coding-related endpoints, which indicates a versioned API contract rather than a fixed end-user app surface.[10]
A separate normalized-data endpoint page makes another distinction from Freed’s public materials. Nabla documents extraction of ICD-10 and LOINC-coded information from Nabla-generated notes in a FHIR-compliant structure, with conditions and observations returned as structured outputs.[12] That does not mean every buyer needs normalized data, but it does indicate that Nabla’s public documentation is stronger on builder-facing, standards-oriented output than on self-serve browser EHR push examples.[12]
The overlap is real, but the public evidence is not symmetrical
There is some genuine overlap. Both products publicly describe transcription or ambient capture, structured clinical note generation, dictation-related functions, and some form of EHR write-back or integration.[3][4][9][11] Both also describe coding-related capabilities in public materials, although through different packaging and documentation styles.[3][4][10][11]
Still, the evidence is not symmetrical enough for a clean checklist match. Freed’s public pages are much clearer about clinician-facing steps, template learning from edits, document upload into chat, prior-visit retrieval, and named browser EHR flows.[3][5][7] Nabla’s public pages are much clearer about API versioning, endpoint-level functions, FHIR-normalized outputs, and embedding inside an EHR or EHR vendor product.[9][10][11][12]
That asymmetry means a strict feature-by-feature verdict would overstate what the public record can support. For example, Nabla’s EHR page says documentation, dictation, and coding flow into the chart, but the cited page does not enumerate field-level routing by EHR the way Freed’s Athenahealth and eClinicalWorks guides do.[11] Conversely, Freed’s public pages describe diagnosis support and chart transfer, but the public documentation does not document a FHIR-normalized extraction service comparable to Nabla’s normalized-data endpoint.[2][12]
Different customer settings are implied by the documentation
Freed’s materials fit a buyer who wants a ready-to-use documentation application. The FAQs describe signing up on the website, capturing a conversation, editing the note, copying content, or using integrations, and the help articles explain setup through the Chrome browser and extension.[7][1] Its feature pages also mention group-oriented functions such as shared patients, medical assistant seats, and admin dashboards, but they are still presented within a clinician application context rather than as a developer platform.[4]
Nabla’s materials point to at least two customer settings. One is the provider organization or health system that wants an AI layer inside an existing EHR workflow.[11] The other is the software company or EHR vendor that wants to build on the Core API or embed Nabla Connect.[9][11] The public pages therefore suggest a broader range of deployment patterns, but they do not by themselves show that all Nabla capabilities are equally available in every deployment mode.[9][11]
That mismatch affects comparison framing. If a medical group is evaluating an out-of-the-box documentation tool for clinicians this quarter, Freed’s public documentation is closer to that buying motion.[1][3][7] If an EHR vendor or digital health platform wants to add medical transcription, note generation, and normalized data through an integration layer or API, Nabla’s public materials are more directly aligned to that scenario.[9][11][12] Neither observation establishes superiority; it only reflects what each company has documented publicly.
What the public documents cannot settle
Several practical buying questions remain open on the public information. Publicly, it is hard to establish like-for-like implementation burden between Freed’s direct product and Nabla’s embedded or API-led models.[1][9][11] It is also unclear how much configuration, governance, or internal technical support a specific deployment would require in a real organization, because the documentation addresses different audiences and scopes.[7][11]
The documents also do not support a broad winner on EHR coverage. Freed names supported browser workflows and documents two-way Athenahealth support on one public page, plus extension and direct connection paths for Athenahealth and eClinicalWorks.[2][6][8] Nabla names several EHR brands on its EHR page, but the supplied page does not break out per-EHR field mapping or browser-style setup instructions.[11] That leaves open whether a buyer’s target workflow depends on a named deployment already live, a custom integration, or a roadmap conversation.
Finally, the evidence cannot settle quality or outcome questions. Freed makes public statements about security, compliance, and internal checks, and Nabla’s pages describe healthcare-grade infrastructure and versioned APIs.[3][10][11] But the supplied comparison brief does not include head-to-head tests, independent validation across the same environments, or enough detail to rank note quality, implementation speed in practice, coding performance, or clinician satisfaction for a particular specialty. A procurement team would need product demos, references, and workflow-specific validation to answer those questions.
Sources & references
- EHR Push | Freed Scribe Help Center · help.getfreed.ai
- Scribe EHR Integration, AthenaHealth, Practice Fusion, & More · www.getfreed.ai
- How Freed Works | AI-Powered Medical Notes That Clinicians Trust · www.getfreed.ai
- Freed features: notes, coding, letters, and EHR push · www.getfreed.ai
- Introducing: AI Chat with Clinical Evidence | Freed Scribe Help Center · help.getfreed.ai
- EHR push for eClinicalWorks | Freed Scribe Help Center · help.getfreed.ai
- Frequently Asked Questions | Freed Scribe Help Center · help.getfreed.ai
- EHR push for Athenahealth | Freed Scribe Help Center · help.getfreed.ai
- Introduction | Nabla Core API documentation · docs.nabla.com
- Changelog | Nabla Core API documentation · docs.nabla.com
- Nabla · EHR Integration · www.nabla.com
- Generate normalized data from a note | Nabla Core API documentation · docs.nabla.com