Nanonets
Nanonets provides AI agents for enterprise document processing, extracting structured data from invoices and similar documents, automating a stated 94% of complex processes while routing low-confidence cases to human review.
27 fields evidenced · 32 with no public information. Every value below links to the document it came from and the date we checked it.
4 of those are flagged for review: our own check found language in the cited source that may address the field. An editor has not adjudicated them yet.
7 of the evidenced values quote wording our own later capture of the same page no longer contains. Marked below, awaiting an editor.
Also appears in
Trust gap
How this is scoredAre scope, spend, and authority enforced while the agent runs?
Low-confidence extractions are routed to human review, a confidence-threshold control at run time; no evidence of enforced scope, spend or authority limits.
Judged Aug 3, 2026; evidence newer. We have since re-read Runtime governance, most recently on Sep 2, 2026. No one has re-scored this dimension against it — the score above stands, but it is older than the evidence below it. How current a score is
How this gap gets closed →Is there a tamper-evident record of what it actually did?
No public evidence found on the page checked.
Judged Aug 3, 2026; evidence newer. We have since re-read Tamper-evident log, most recently on Sep 2, 2026. No one has re-scored this dimension against it — the score above stands, but it is older than the evidence below it. How current a score is
How this gap gets closed →Are compliance obligations attached per engagement?
No public evidence found on the page checked.
Judged Aug 3, 2026; evidence newer. We have since re-read Regulatory alignment, most recently on Sep 2, 2026. No one has re-scored this dimension against it — the score above stands, but it is older than the evidence below it. How current a score is
How this gap gets closed →Does payment depend on a verified result?
No public evidence found on the page checked.
Judged Aug 3, 2026; evidence newer. We have since re-read Outcome-based pricing and Settlement mechanism, most recently on Sep 2, 2026. No one has re-scored this dimension against them — the score above stands, but it is older than the evidence below it. How current a score is
How this gap gets closed →Can the deployment be insured, and is the customer indemnified?
No public evidence found on the page checked.
Judged Aug 3, 2026; evidence newer. We have since re-read Insurance available, Insurance carriers, Coverage limits and Indemnification, most recently on Sep 2, 2026. No one has re-scored this dimension against them — the score above stands, but it is older than the evidence below it. How current a score is
How this gap gets closed →Assurance
- Insurance available
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Own liability cover
- no public informationnanonets.com · checked Aug 14, 2026 · source did not state this
- Insurance carriers
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Coverage limits
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Indemnification
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Liability cap
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Audit trail
- Audit logs stated alongside SSO, private deployment and data residency
“SSO, audit logs, private deployment, data residency”
Not in our latest capture. This quote was on the page when we filed it. We captured the same page again on Sep 2, 2026 and it was not there. The value above has not been re-checked by an editor yet. How we re-check citations
nanonets.com · checked Aug 2, 2026 - Tamper-evident log
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Explainability
- States every agent run is traceable, auditable, and explainable, scoped by RBAC, implying decisions can be traced to their source rules.
“Every agent run in the context graph is traceable, auditable, explainable, and runs on RBAC scopes.”
nanonets.com · checked Sep 2, 2026 - Runtime governance
- Enforces per-team budgets, API quotas, and rate limits at runtime with real-time cost dashboards, alongside RBAC scoping of agent authority.
“Usage quotas & rate limits Per-team budgets and API quotas with real-time cost dashboards. No surprises.”
nanonets.com · checked Sep 2, 2026 - Permission scopes
- Role-based access with fine-grained permissions scoped per workspace, agent and data source
“Fine-grained permissions per workspace, agent, and data source.”
Not in our latest capture. This quote was on the page when we filed it. We captured the same page again on Sep 2, 2026 and it was not there. The value above has not been re-checked by an editor yet. How we re-check citations
nanonets.com · checked Aug 2, 2026 - Compliance certifications
- SOC 2 Type II claimed, alongside GDPR and ISO 27001. Self-declared; no certificate or auditor named.
“SOC 2 Type II Certified for ongoing security controls. GDPR Aligned with EU data protection laws. ISO 27001 Compliant with global ISMS standards.”
nanonets.com · checked Aug 2, 2026 - Regulatory alignment
- States alignment with GDPR / EU data-protection law requirements.
“GDPR Aligned with EU data protection laws.”
nanonets.com · checked Sep 2, 2026 - Outcome-based pricing
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Settlement mechanism
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Asset custody
- no public informationnanonets.com · checked Aug 14, 2026 · source did not state this
- Dispute process
- no public information
Flagged for review — the source below contains language that may address this field. Not yet checked by an editor. How we check absences
nanonets.com · checked Sep 2, 2026 - SLA terms
- no public information
Flagged for review — the source below contains language that may address this field. Not yet checked by an editor. How we check absences
nanonets.com · checked Sep 2, 2026 - Evaluation coverage
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Self-reported performance
- no public informationnanonets.com · checked Aug 14, 2026 · source did not state this
Agency
- Autonomy level
- Vendor states 94% of complex, document-heavy processes are automated, with 6% left for human judgment (self-reported)
“94% of complex, document-heavy processes, automated. 6% left for human judgment.”
Not in our latest capture. This quote was on the page when we filed it. We captured the same page again on Sep 2, 2026 and it was not there. The value above has not been re-checked by an editor yet. How we re-check citations
nanonets.com · checked Aug 2, 2026 - Human oversight
- Low-confidence extractions are routed to human review
“Human review on low confidence”
Not in our latest capture. This quote was on the page when we filed it. We captured the same page again on Sep 2, 2026 and it was not there. The value above has not been re-checked by an editor yet. How we re-check citations
nanonets.com · checked Aug 2, 2026 - Goal complexity
- Describes its agents as completing work end-to-end -- reading inputs, applying business rules, and posting decisions -- a multi-step process, not a single action.
“Agents complete the work end-to-end — reading inputs, applying your rules, and posting decisions into your systems of record.”
nanonets.com · checked Sep 2, 2026 - Action space
- Extracts structured data from invoices, purchase orders, bills of lading, and other business documents
“Extract structured data from the documents behind your processes, like invoices, POs, bills of lading.”
Not in our latest capture. This quote was on the page when we filed it. We captured the same page again on Sep 2, 2026 and it was not there. The value above has not been re-checked by an editor yet. How we re-check citations
nanonets.com · checked Aug 2, 2026 - Operating environment
- Operates embedded within enterprise systems (ERP/finance, procurement, supply chain, healthcare) rather than as a standalone chat interface.
“Agents run on context graphs that connect every document, system, rule. Your workflows across finance, operations, healthcare close without a human.”
nanonets.com · checked Sep 2, 2026 - Initiative
- States confidence scores determine when an agent proceeds autonomously versus escalating to a human, implying it can act without a fresh prompt each time.
“Confidence scores decide when the agent acts autonomously and when it asks a human.”
nanonets.com · checked Sep 2, 2026
Safety
- Safety evaluations
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Red teaming
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Safety policy
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Usage restrictions
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Model or system card
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Incident reporting
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Third-party evaluations
- no public information
Flagged for review — the source below contains language that may address this field. Not yet checked by an editor. How we check absences
nanonets.com · checked Sep 2, 2026 - Data handling
- States customer data is encrypted with AES-256 at rest and TLS 1.3 in transit, with support for customer-managed keys (BYOK).
“Encryption at rest & in transit AES-256 at rest, TLS 1.3 in transit. BYOK supported for customer-managed keys.”
nanonets.com · checked Sep 2, 2026
Practicality
- Pricing model
- Offers a free tier with $50 in trial credits (no card required) suggesting a usage/credit-based pricing structure alongside enterprise plans.
“Get started for free $50 free credits, no card required.”
nanonets.com · checked Sep 2, 2026 - Price point
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Availability
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Deployment options
- Offers private VPC, single-tenant cloud, or on-premise deployment options.
“Private deployment VPC, single-tenant cloud, or on-prem. Choose your infra and network policies.”
nanonets.com · checked Sep 2, 2026 - Integrations
- Delivers extracted data to ERP, CRM, or database systems (example shown delivering to SAP)
“Delivered to your ERP, CRM, or database”
Not in our latest capture. This quote was on the page when we filed it. We captured the same page again on Sep 2, 2026 and it was not there. The value above has not been re-checked by an editor yet. How we re-check citations
nanonets.com · checked Aug 2, 2026 - Supported regions
- Lets customers pin data processing to specific regions (US, EU, or APAC) so data stays within that geographic boundary.
“Data residency Pin data processing to US, EU, APAC regions. Customer data never leaves your boundary.”
nanonets.com · checked Sep 2, 2026 - Support model
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
Foundation models
- Base models
- Uses its own proprietary model, Nanonets OCR-3, for the core data-extraction task, claimed as the most accurate such model available.
“Nanonets OCR-3, our proprietary model, is the most accurate data extraction model in the world, making the rest of your pipeline safe to automate.”
nanonets.com · checked Sep 2, 2026 - Model provider
- Supplies its own in-house model (Nanonets OCR-3) rather than relying on an external model provider for core extraction.
“Nanonets OCR-3, our proprietary model, is the most accurate data extraction model in the world, making the rest of your pipeline safe to automate.”
nanonets.com · checked Sep 2, 2026 - Model swappable
- no public information
Flagged for review — the source below contains language that may address this field. Not yet checked by an editor. How we check absences
nanonets.com · checked Sep 2, 2026 - Open weights
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Fine-tuning
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Context window
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
Ecosystem
- Protocols supported
- Supports interoperability via the Model Context Protocol (MCP) alongside native retrievers and a REST/GraphQL API, plus integration with major agent platforms.
“Point the agents your teams already run at Nanonets through an MCP server, native retrievers, or plain REST/GraphQL API.”
nanonets.com · checked Sep 2, 2026 - Tool use
- Connects to external systems and communication tools (Slack, Zoom, Teams, email) as well as ERP/finance systems to extract information for agent decisions.
“Connect your systems Nanonets connects to Slack, Zoom, Teams, email, all the places where your team's unwritten rules and precedents live, and extracts them.”
nanonets.com · checked Sep 2, 2026 - Multi-agent
- Shows multiple distinct named agents (e.g. Freight Audit, AP Clerk, Regional Manager, Business Analyst) participating together within a single workflow run.
“Other Agents in this run AP Clerk Agent Scoped access US West · Central Regional Manager Agent Read-only access Americas · Europe Business Analyst Agent Configure access No integrations”
nanonets.com · checked Sep 2, 2026 - API access
- Exposes a REST/GraphQL API for programmatic access, in addition to an MCP server and native retrievers.
“REST/GraphQL API”
nanonets.com · checked Sep 2, 2026 - Open source
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Marketplace presence
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
Impact
- User base
- Reports usage by roughly 35% of the Fortune 500 and over 10,000 enterprises overall.
“Used by 35% of the Fortune 500 and 10,000+ enterprises”
nanonets.com · checked Sep 2, 2026 - Deployment scale
- States customer deployments typically reach production within weeks rather than quarters, aided by forward-deployed engineers.
“Our team of forward-deployed engineers ensure that your AI projects go live in weeks, not quarters.”
nanonets.com · checked Sep 2, 2026 - Target sectors
- Named use cases include invoice processing, order processing, shipment reconciliation, and loan processing
“Invoice processing Order processing Shipment reconciliation Loan processing”
Not in our latest capture. This quote was on the page when we filed it. We captured the same page again on Sep 2, 2026 and it was not there. The value above has not been re-checked by an editor yet. How we re-check citations
nanonets.com · checked Aug 2, 2026 - High-risk domains
- Explicitly serves healthcare revenue-cycle management and automates insurance claims adjudication, both high-risk/regulated domains.
“+ Healthcare RCM”
nanonets.com · checked Sep 2, 2026 - Documented incidents
- no public informationnanonets.com · checked Sep 2, 2026 · source did not state this
- Market recognition
- no public informationnanonets.com · checked Aug 14, 2026 · source did not state this
Compared with
Side-by-side on the fields where Nanonets and the other entry are both on record and actually differ.