llms.txt
Compare with other sites' llms.txt files →# Corti Labs
> Frontier open-weight models, hardened and hosted in Europe. One OpenAI-compatible API on EU infrastructure we own. Free credits to start.
Corti builds frontier healthcare AI: open-weight models, speech, coding, and agents on sovereign EU infrastructure. Prefer Markdown mirrors (`.md`) and this file for machine-readable context.
## Optional
- [Full content map](https://corti.ai/llms-full.txt): Expanded catalogue with agent and connector listings plus recent article Markdown.
- [API docs llms.txt](https://docs.corti.ai/llms.txt): Product documentation for models and APIs.
## Platform
- [Corti Models](https://corti.ai/): Frontier open-weight models, hardened and hosted in Europe. One OpenAI-compatible API on EU infrastructure we own. Free credits to start.
- [Models overview](https://corti.ai/models): Frontier open-weight models, hardened and hosted in Europe. One OpenAI-compatible API on EU infrastructure we own. Free credits to start.
- [Speech to text](https://corti.ai/speech-to-text): Clinical-grade speech-to-text for ambient and dictation workflows.
- [Medical coding](https://corti.ai/medical-coding): Agentic medical coding across ICD-10 and CPT with evidence.
- [Text generation](https://corti.ai/text-generation): Healthcare-native text generation for clinical documentation.
- [Symphony](https://corti.ai/symphony): Orchestration layer for speech, documentation, coding, and agents.
- [Sovereign Cloud](https://corti.ai/sovereign-cloud): Run Corti Models in an EU-operated cloud or on your own hardware. Same APIs, residency you can defend, no US hyperscaler on the request path.
- [Compliance](https://corti.ai/compliance): How Corti Labs handles encryption, residency, access, monitoring, and audit for regulated workloads. Review live controls in the Trust Center.
- [Agentic Framework](https://corti.ai/agentic-framework): Create agents with typed connectors, persistent contexts, and A2A messaging. Plan, execute, and trace multi-agent workflows on EU or US infrastructure.
- [Pricing](https://corti.ai/pricing): Pay-as-you-go and enterprise pricing for Corti's healthcare-native APIs. Start for free, scale to custom enterprise plans.
- [About](https://corti.ai/about): Corti is the frontier AI lab hardened in healthcare and built for critical industries.
- [Contact](https://corti.ai/contact): Talk to an AI expert about running frontier models on sovereign EU infrastructure - deployment, data residency, procurement and pricing for your team.
## Research
- [Research](https://corti.ai/research): Corti's peer-reviewed research publications on clinical AI
## Newsroom
- [Newsroom](https://corti.ai/newsroom): Press releases and company announcements
## Agents
- [Agents index](https://corti.ai/agents): Agent Blueprints and the connectors you can bundle into an agent
- [Medical Coding Agent](https://corti.ai/agents/medical-coding-icd-10-cpt-agent): Generate accurate medical codes from clinical context
- [Clinical Education Agent](https://corti.ai/agents/clinical-education-agent): Access evidence-based clinical explanations
- [Clinical Guidelines Agent](https://corti.ai/agents/clinical-guidelines-agent): Evaluate patient care against clinical guidelines
- [Nursing Shift Handoff Agent](https://corti.ai/agents/nursing-shift-handoff-agent): Consolidate patient information for easy handoffs
- [Clinical Documentation Improvement (CDI) Agent](https://corti.ai/agents/clinical-documentation-improvement-cdi-agent): Review clinical documentation for accuracy gaps
- [Note Completeness Agent](https://corti.ai/agents/note-completeness-agent): Ensure clinical note completeness in real-time
- [Triage and Initial Assessment Agent](https://corti.ai/agents/triage-and-initial-assessment-agent): Automate clinical triage and risk stratification
- [Referral Generator Agent](https://corti.ai/agents/referral-generator-agent): Generate clinician-to-clinician referral letters
- [Prior Authorization Agent](https://corti.ai/agents/prior-authorization-agent): Generate evidence-based prior authorization letters
- [Denial Appeals Agent](https://corti.ai/agents/denial-appeals-agent): Appeal denials with evidence-backed responses
- [Medication Reconciliation Agent](https://corti.ai/agents/medication-reconciliation-agent): Review medication against clinical records
- [Patient Discharge Education Agent](https://corti.ai/agents/patient-discharge-education-agent): Convert clinical content into patient-friendly outputs
- [Surgical Registry Intelligence Agent](https://corti.ai/agents/surgical-registry-intelligence-agent): Automate surgical quality registry data entry
- [Code Validation Agent](https://corti.ai/agents/code-validation-agent): Validate ICD-10-CM, ICD-10-PCS, and CPT code sets
- [Diagnostic Entity Extractor Agent](https://corti.ai/agents/diagnostic-entity-extractor-agent): Extract codeable diagnoses and assign codes directly
- [ICU Admission Summary Agent](https://corti.ai/agents/icu-admission-summary-agent): Generate structured ICU admission summaries
- [Compliance Guardrail Agent](https://corti.ai/agents/compliance-guardrail-agent): Validate code sets against payer or organizational rulesets
- [ICD-10 Index Navigator Agent](https://corti.ai/agents/icd-10-index-navigator-agent): Go from clinical terms to candidate codes for coder review
- [Procedure Entity Extractor Agent](https://corti.ai/agents/procedure-entity-extractor-agent): Extract codeable procedures and assign codes directly
- [Rule Explainer Agent](https://corti.ai/agents/rule-explainer-agent): Understand code selection reasoning
## Connectors
- [Medical Coding](https://corti.ai/connectors/coding): Assign accurate medical codes from clinical input
- [Web Search](https://corti.ai/connectors/web-search): Pull the latest content from the internet
- [Clinical Trials](https://corti.ai/connectors/clinical-trials): Search and access structured clinical trial data
- [Interviewing](https://corti.ai/connectors/interviewing): Run structured, adaptive questionnaires
- [DrugBank](https://corti.ai/connectors/drugbank): Access comprehensive, up-to-date drug intelligence
- [Posos](https://corti.ai/connectors/posos): Access the Posos medication database
- [PubMed](https://corti.ai/connectors/pubmed): Retrieve research directly from PubMed
- [Medical Calculator](https://corti.ai/connectors/medical-calculator): Perform reliable medical calculations
- [CCAM Coding (coming soon)](https://corti.ai/connectors/ccam-coding): Assign accurate CCAM medical codes from clinical input
- [CIM-10-FR Inpatient Coding (coming soon)](https://corti.ai/connectors/cim-10-fr-inpatient-coding): Assign accurate CIM-10-FR inpatient medical codes
- [CIM-10-FR Outpatient Coding (coming soon)](https://corti.ai/connectors/cim-10-fr-outpatient-coding): Assign accurate CIM-10-FR outpatient medical codes
- [CPOE Recommendations (coming soon)](https://corti.ai/connectors/cpoe-recommendations): Surface the right orders and order sets
- [Checklist (coming soon)](https://corti.ai/connectors/checklist): Monitor care protocol adherence
- [ICD-10-CM Coding](https://corti.ai/connectors/coding-expert-icd-10-cm): Assign accurate ICD-10-CM medical codes
- [ICD-10-GM Inpatient Coding (coming soon)](https://corti.ai/connectors/icd-10-gm-inpatient-coding): Assign accurate ICD-10-GM inpatient medical codes
- [ICD-10-GM Outpatient Coding (coming soon)](https://corti.ai/connectors/icd-10-gm-outpatient-coding): Assign accurate ICD-10-GM outpatient medical codes
- [ICD-10-INT Coding](https://corti.ai/connectors/coding-expert-icd-10-int): Assign accurate ICD-10-INT medical codes
- [ICD-10-PCS Coding](https://corti.ai/connectors/coding-expert-icd-10-pcs): Assign accurate ICD-10-PCS medical codes
- [ICD-10-UK Coding](https://corti.ai/connectors/coding-expert-icd-10-uk): Assign accurate ICD-10-UK medical codes
- [Memory](https://corti.ai/connectors/memory): Retrieve precise information from complex data
- [OPCS-4 Coding (coming soon)](https://corti.ai/connectors/opcs-4-coding): Assign accurate OPCS-4 medical codes from clinical input
- [OPS Coding (coming soon)](https://corti.ai/connectors/ops-coding): Assign accurate OPS medical codes from clinical input
## Industries
- [Public sector](https://corti.ai/industries/public-sector): Harness AI without giving away citizen data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Central government](https://corti.ai/industries/central-government): Harness AI capabilities without giving away your data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Regions & municipalities](https://corti.ai/industries/regions-municipalities): Harness AI without giving away resident data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Emergency services](https://corti.ai/industries/emergency-services): Harness AI without giving away incident data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Life sciences](https://corti.ai/industries/life-sciences): Harness AI without giving away research data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Pharma](https://corti.ai/industries/pharma): Harness AI without giving away trial data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Clinical research & CROs](https://corti.ai/industries/clinical-research): Harness AI without giving away patient data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Financial services](https://corti.ai/industries/financial-services): Harness AI without giving away client data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Banking](https://corti.ai/industries/banking): Harness AI without giving away customer data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Insurance](https://corti.ai/industries/insurance): Harness AI without giving away policyholder data. Modernize legacy systems and build agents with models you own, every action auditable.
- [Legal](https://corti.ai/industries/legal): Add AI features to legal software and keep client data in region. Modernize your platform and build agents with models you own.
## Resources
- [Blog](https://corti.ai/blog): The latest from Corti.
- [Blog](https://corti.ai/blog): The latest from Corti — analysis, announcements, and customer perspectives.
- [Guides](https://corti.ai/guides): Practical guides for building with and evaluating Corti.
- [Builder Series](https://corti.ai/builder-series): Conversations with the teams building on Corti.
## Latest articles
- [What's new in our Agentic Framework](https://corti.ai/guides/what-is-new-in-our-agentic-framework.md): Six new features in the Corti Agentic Framework: connectors and local tools, token streaming, inline citations, a feedback API, OpenInference traces, and open-source evals. What each one does and the
- [October Wrapped](https://corti.ai/blog/october-wrapped.md): Frontier AI is advancing fast. But building powerful models is only part of the challenge. Putting them to work, reliably and at scale, is what comes next.
- [Healthcare AI is entering the multi-agent era. The infrastructure must keep up.](https://corti.ai/blog/healthcare-ai-is-entering-the-multi-agent-era-the-infrastructure-must-keep-up.md)
- [How to code as a clinician with no coding experience](https://corti.ai/guides/how-to-code-as-a-clinician-with-no-coding-experience.md): You don't need to be a developer to build working clinical software. This guide walks clinicians through agentic coding with OpenCode and Corti models, from free account setup to a first working demo.
- [Symphony for Text Generation: Introducing a new evaluation benchmark for clinical note quality](https://corti.ai/blog/symphony-for-text-generation-benchmark-clinical-note-quality.md): We built a controlled, multilingual way to measure what AI-generated clinical notes keep, leave out, and add. Then we used it to test Symphony for Text Generation against Heidi and Tandem Health.
- [Six use cases enabled by Symphony for Medical Coding SNOMED CT](https://corti.ai/guides/six-things-you-can-build-on-snomed-ct.md): SNOMED CT lets software read clinical data by meaning, not by code, but manual coding never scaled. With Symphony for Medical Coding, turning notes into structured concepts is now one API call. Here a
- [Clinical AI Innovation Deep Dive: Exploring the fragmented data timeline problem in EHRs](https://corti.ai/blog/clinical-ai-innovation-deep-dive-exploring-the-fragmented-data-timeline-problem-in-ehrs.md): Patient history in the EHR grew from 1,700 words in 2006 to nearly 60,000 in 2022. Why clinicians still build the patient timeline by hand, and how to fix it.
- [Symphony for Medical Coding now supports SNOMED CT](https://corti.ai/blog/symphony-for-medical-coding-now-supports-snomed-ct.md): Symphony for Medical Coding now supports SNOMED CT, giving developers concept-level coding across the International edition and seven national editions through a single API call.
- [The structured data problem in healthcare](https://corti.ai/blog/the-structured-data-problem-in-healthcare.md): Unstructured data in healthcare hides what clinicians need. Where SNOMED CT coding helps, where it falls short, and how semantic search fills the gap.
- [What if there was a search engine for your patient's history?](https://corti.ai/blog/what-if-there-was-a-search-engine-for-your-patients-history.md): EHR search leaves clinicians scrolling. What a clinical search engine for patient history needs, from semantic retrieval and provenance to role-aware results.
- [A Guide to building ambient scribing that handles the physical exam](https://corti.ai/guides/a-guide-to-building-ambient-scribing-that-handles-the-physical-exam.md): Ambient scribing in somatic medicine works well for most of the consultation, but the physical examination needs special attention. This guide explains where the challenges are and what clinicians and
- [Why the next era of AI scribes needs an entirely new architecture](https://corti.ai/blog/why-the-next-era-of-ai-scribes-needs-an-entirely-new-architecture.md)
- [Introducing Corti ADVANCE: Bridging the gap between AI ambition and AI deployment](https://corti.ai/blog/introducing-corti-advance-bridging-the-gap-between-ai-ambition-and-ai-deployment.md): ADVANCE pairs Corti's production-validated agent patterns and API platform with forward deployed engineers, helping our customers take AI initiatives from pilot to production on sovereign infrastructu
- [Welcome to medical speech, ElevenLabs](https://corti.ai/blog/welcome-to-medical-speech-elevenlabs.md): ElevenLabs used Corti’s medical speech benchmarks to test Scribe v2 Medical. We look at what the results show and why we’re rerunning the Symphony comparison.
- [AI is only as good as the workflow around it](https://corti.ai/builder-series/ai-is-only-as-good-as-the-workflow-around-it.md): Voice Technologies and Corti are integrating clinical AI, ambient transcription and speech recognition into everyday NHS workflows.
- [No European companies are asking for sovereign AI](https://corti.ai/blog/no-european-companies-are-asking-for-sovereign-ai.md): European companies don't buy AI for the flag. They buy the best they're allowed to depend on. Near-frontier open weights and better harnesses are changing the sovereign AI equation.
- [EU data residency: What European healthcare builders need to know](https://corti.ai/guides/eu-data-residency-what-european-healthcare-builders-need-to-know.md): EU healthcare buyers require proof, not claims, that patient data is stored and processed within the EU, including through subprocessors.
- [AI vendor evaluation for US healthtech: The compliance checklist](https://corti.ai/guides/ai-vendor-evaluation-for-us-healthtech-the-compliance-checklist.md): Evaluating an AI vendor for a US healthcare product means navigating a compliance landscape that is sector-specific rather than one-size-fits-all.
- [How to evaluate medical speech-to-text (ASR): WER, CER, and clinical benchmarks](https://corti.ai/guides/how-to-evaluate-automatic-speech-recognition.md): Evaluate medical speech-to-text with WER, CER, and clinical term benchmarks — a practical framework for comparing ASR on your own healthcare audio.
- [What makes a great automated medical coding system: rare codes, evidence, and reviewability](https://corti.ai/guides/what-makes-a-great-automated-medical-coding-system.md): Headline accuracy hides blind spots. What production coding needs: rare-code performance, care-setting rules, evidence spans, and output reviewers can trust.
- [Corti completes ISO 27001, 27017, and 27018 with zero nonconformities](https://corti.ai/blog/corti-completes-iso-audits-with-zero-nonconformities.md): Corti achieves zero nonconformities across ISO 27001, ISO 27017, and ISO 27018 audits, reinforcing independently verified security, cloud governance, and data privacy.
- [Why healthcare needs multi-agent systems instead of one LLM for every clinical task](https://corti.ai/guides/why-healthcare-needs-multi-agent-systems.md): Learn why single-agent AI struggles in healthcare and how Corti’s multi-agent architecture delivers focused expertise, faster execution, and better results.
- [Ambient AI in psychiatry and psychology: Solving clinical documentation challenges](https://corti.ai/guides/ambient-scribing-in-psychiatry-and-psychology.md): Ambient AI in psychiatry and psychology requires domain-specific design; this guide explains what must change to ensure safe deployment.
- [Three pre-charting summarization workflow examples on Corti for development teams](https://corti.ai/guides/pre-charting-summarization-three-use-cases-built-on-corti.md): Three pre-charting workflows on Corti — packed primary care schedules, complex specialty cases, and behavioural health — with summarisation architecture and cost examples.
- [Healthcare AI needs more care, not just more speed](https://corti.ai/builder-series/healthcare-ai-needs-more-care-not-just-more-speed.md): We explore how AI in healthcare can support trust, attention, and human judgment while helping clinicians stay present with patients.
- [AI-native RCM companies: who is building, where gaps remain, and why builders need APIs](https://corti.ai/guides/ai-native-entrants-in-rcm.md): Who is building AI-native RCM — and where applications leave gaps. Why product teams need coding APIs they can embed, not another enterprise procurement cycle.
- [Build vs buy healthcare AI agents: a developer decision guide](https://corti.ai/guides/ai-agents-solving-the-build-vs-buy-dilemma-for-developers.md): Build, buy, or something in between? A developer guide to healthcare AI agent infrastructure — and where purpose-built platforms save months of work.
- [Context engineering for multi-agent healthcare AI: how the Corti Agentic Framework handles context](https://corti.ai/guides/context-engineering-for-high-stakes-ai-deployment.md): How the Corti Agentic Framework handles memory, semantic retrieval, and strict data isolation — so the right clinical context reaches the right agent every time.
- [Corti pre-built Experts: specialized agents for coding, clinical references, and drug lookups](https://corti.ai/guides/the-power-of-specialized-agents-corti-agentic-framework-pre-built-experts.md): Pre-built Corti Experts for coding, clinical references, and drug lookups — MCP-compatible agents you can compose, extend, or replace with your own tools.
- [Healthcare agent infrastructure for production: compliance, audit trails, and scale at volume](https://corti.ai/guides/healthcare-agent-infrastructure-purpose-built-for-production.md): Running healthcare agents in production needs more than a prototype — access control, audit trails, tracing, and scale. What purpose-built i