Model Choice
Members can select enabled first-party model and reasoning options; admins control model availability and automatic routing.
2 / 5
Multiple first-party model variants can be selected.
Evidence-led enterprise AI market mapSource
ChatGPT Enterprise
OpenAI
SaaS95% verified
WebsiteVerified 2026-07-16
Calculated axis scores
Weighted rubric averages. No axis score is stored in YAML.
Configurability
3.00
out of 5.00
Enterprise readiness
3.82
out of 5.00
Rubric scorecard
Every score, reached definition, rationale, and source.
Configurability
Custom Agents
Workspace Agents combine instructions, files, skills, models, apps, schedules, permissions, sharing, controlled actions, and publishing.
4 / 5
Organization-managed agents support permissions, sharing, approved data, and controlled tools.
Multi-Agent Support
Official ChatGPT Enterprise sources reviewed do not document agent-to-agent delegation or configurable agent teams.
0 / 5
Not verified — excluded from scoring.
No proof
MCP / Tool Integrations
Remote MCP apps support authenticated actions, admin vetting and publishing, RBAC, per-action controls, update review, approvals, and compliance logging.
5 / 5
Adds governed discovery, allowlists, granular permissions, auditability, and lifecycle controls for remote tools.
Workflow Automation
Workspace Agents run long, multi-step workflows through configurable tools, schedules, API triggers, and Slack deployments.
3 / 5
Multi-step workflows with triggers and configurable actions.
APIs & SDKs
Scoped Workspace Agent access tokens support API-triggered automation; Compliance APIs expose enterprise workspace logs and state.
4 / 5
Enterprise APIs cover agents, tools, knowledge, administration, or automation with service authentication.
Knowledge / RAG
Company Knowledge retrieves from synced sources with citations, source-permission enforcement, OAuth, RBAC, and group-level admin controls.
4 / 5
Permission-aware enterprise sources support sync, citations, and admin controls.
Evaluation Framework
Builders can preview and manually test agents with sample prompts and review agent analytics; repeatable datasets or graders are not documented.
2 / 5
Manual feedback or simple thumbs-up/down analytics.
Prompt Customization
Workspace Instructions and reusable agent templates provide saved instruction patterns; a separate governed prompt-asset lifecycle is not documented.
2 / 5
Saved prompts or reusable instruction templates.
Extensibility
Organizations can package plugins and custom MCP apps, distribute them through a directory, publish after review, and administer role/action access.
4 / 5
Developers can package, distribute, and administer extensions across an organization.
Deployment Flexibility
Enterprise provides a dedicated organization workspace plus customer-selectable storage regions and eligible US/EU inference residency.
3 / 5
Dedicated cloud, region selection, or private networking is available.
Enterprise readiness
SSO
Enterprise supports SAML and OIDC SSO with verified-domain enforcement; one organization ID supports one IdP configuration.
3 / 5
Standards-based SAML or OIDC SSO for enterprise workspaces.
SCIM
SCIM automates provisioning and deprovisioning, syncs groups, and lets synced groups drive role assignment and access.
4 / 5
SCIM groups and role or team mapping with automated lifecycle management.
RBAC
Owners define custom roles and assign them to groups for granular access to GPTs, projects, apps, agents, and workspace features.
4 / 5
Custom roles or granular permissions cover data, agents, tools, and administration.
Audit Logs
Immutable append-only logs and state APIs integrate with SIEM, DLP, and eDiscovery; logs retain 30 days and support continuous export.
4 / 5
Export, API, or SIEM streaming provides broad documented event coverage and retention controls.
Compliance
Assurance includes SOC 2 Type II, ISO 27001/27017/27018/27701, GDPR support, CSA STAR alignment, Trust Portal evidence, and a FedRAMP Moderate Enterprise variant.
4 / 5
Multiple relevant certifications or attestations cover enterprise operations and regulated workloads.
Data Residency
Eligible Enterprise customers select among ten storage-at-rest regions for customer content; US and Europe also support inference residency.
3 / 5
Customer-selectable storage region for primary product data.
Governance
Central policies cover models, agents, GPTs, data apps, plugins, tools, publishing, RBAC, action approval gates, and monitoring.
4 / 5
Central policy covers agents, connectors, data, models, and lifecycle with approval workflows.
Admin Controls
Global and workspace consoles manage multiple workspaces, delegated roles, domains, SSO, groups, members, features, policies, and bulk provisioning.
4 / 5
Delegated administration, groups, domains, policies, and bulk controls are supported.
Analytics
Role-scoped dashboards and CSV exports report adoption, users, GPTs, projects, tools, groups, trends, benchmarks, impact, usage, budgets, and spend controls.
4 / 5
Export or API supports custom reporting, cost allocation, and trend analysis.
Enterprise Support
Enterprise offers 24/7 priority support, SLAs, dedicated onboarding, ongoing account management, AI advisors, custom legal terms, and security review.
4 / 5
Named technical or account contacts, priority SLAs, and implementation or security support.
Security Controls
Controls include no training by default, custom retention, AES-256/TLS, IP allowlisting, external BYOK across major clouds, and DLP/eDiscovery/SIEM integration.
4 / 5
Customer-managed retention, private networking or IP controls, key options, and advanced threat protections.
Calculation
For each axis: sum(score × rubric weight) ÷ sum(weights). Any result without verified evidence is constrained to score 0 by schema.