Answer: Open Policy & security → Models, configure Workspace default, then use Roles for role-specific settings; review role overrides for each product separately, since Allow medium/high/extra high effort and Allow Pro models apply to Chat while Allow fast mode and Allow Ultra Fast mode apply to Work & Codex. Status: Confirmed Entry: tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits Revision: 1 Digest: sha256:cc83a2677ec64ecb (changes whenever the content changes) Timeline: Created: 2026-11-13T03:50:07Z Last changed: 2026-11-13T03:50:07Z Last verified: 2026-11-20T13:38:14Z Next check: 2026-11-21T13:38:14Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of Navigation path | Policy & security → Models | [1] | 2026-11-20 Configuration order | Workspace default | [1] | 2026-11-20 Chat effort/pro model controls | Allow medium/high/extra high effort | [1] | 2026-11-20 Work & Codex speed controls | Allow fast mode | [1] | 2026-11-20 Separate review guidance | Review role overrides for each product separately. | [1] | 2026-11-20 Sources: [1] OpenAI. "Best practices for ChatGPT Edu workspace configuration in higher education". https://learn.chatgpt.com/docs/enterprise/higher-education-roles-and-permissions.md Role: primary. Accessed: 2026-11-20T13:38:14Z Quote: "Open **Policy & security → Models**." (Supports: Navigation path) Quote: "Configure **Workspace default**, then use **Roles** for role-specific settings." (Supports: Configuration order) Quote: "**Allow medium/high/extra high effort** and **Allow Pro models** apply to Chat" (Supports: Chat effort/pro model controls) Quote: "**Allow fast mode** and **Allow Ultra Fast mode** apply to Work & Codex." (Supports: Work & Codex speed controls) Quote: "Review role overrides for each product separately." (Supports: Separate review guidance) Snapshot SHA-256: b09a82058c4b2c47b73a5041e25ee3946df17e6470a96da31a95b3a2904a8295 (captured 2026-11-13T03:50:05Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-13T03:50:07Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T13:38:14Z. No independent second source is configured for this entry Last successful check: 2026-11-20T13:38:14Z Last attempt: 2026-11-20T13:38:14Z Notes and limitations: Extracted automatically from a single official source (section "Model access by role"). Related: https://latestn.com/tool/chatgpt/admin-and-enterprise/monitor-usage-manage-credits How can I monitor usage and manage credits for ChatGPT Business or Enterprise/Edu? https://latestn.com/tool/chatgpt/admin-and-enterprise/move-plugin-to-github How do I move an existing ChatGPT plugin to GitHub management? https://latestn.com/tool/chatgpt/admin-and-enterprise/multiple-group-membership Can a ChatGPT Enterprise member belong to more than one group? https://latestn.com/tool/chatgpt/admin-and-enterprise/meetings-enterprise-availability Is the Meetings plugin in ChatGPT available for ChatGPT Enterprise? https://latestn.com/tool/chatgpt/admin-and-enterprise/migrate-personal-workspace How do I migrate my personal ChatGPT workspace to a Business or Enterprise workspace? https://latestn.com/tool/chatgpt/admin-and-enterprise/mix-seats-workspace Can I mix standard Enterprise and Financial Services seats in one ChatGPT workspace? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T03:50:07Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits POST: https://latestn.com/dispute/tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How do I set model access and usage limits separately for ChatGPT Edu roles? Revision 1, last verified 2026-11-20T13:38:14Z. https://latestn.com/tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits (accessed 2026-11-21T10:02:46Z). @misc{LatestN_tool_chatgpt_admin_and_enterprise_model_access_and_usage_limits_r1, title = {How do I set model access and usage limits separately for ChatGPT Edu roles?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits}}, note = {Revision 1; digest sha256:cc83a2677ec64ecb; last verified 2026-11-20T13:38:14Z; accessed 2026-11-21T10:02:46Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/chatgpt/admin-and-enterprise/model-access-and-usage-limits