{"id":2888,"date":"2026-09-02T04:46:31","date_gmt":"2026-09-02T04:46:31","guid":{"rendered":"https:\/\/dmarketertayeeb.com\/blog\/github-copilot-september-2026-model-budget-review-changes\/"},"modified":"2026-10-03T16:38:11","modified_gmt":"2026-10-03T16:38:11","slug":"github-copilot-september-2026-model-budget-review-changes","status":"publish","type":"post","link":"https:\/\/dmarketertayeeb.com\/blog\/github-copilot-september-2026-model-budget-review-changes\/","title":{"rendered":"GitHub Copilot September 2026: Model Deprecations, Gemini 3.8 and Signup Changes"},"content":{"rendered":"\n

October 3, 2026 update:<\/strong> For teams reviewing GitHub Copilot’s September 2026 model, access and policy changes, four models retired on October 2. GitHub’s suggested replacements are: Gemini 3.5 Flash and Gemini 3.6 Flash to Gemini 3.8 Flash, Kimi K2.7 Code to Kimi K3, and Claude Opus 4.7 to Claude Opus 5.5, as listed in the October 2 notice<\/a>. GitHub says no manual removal is needed. Check model access and update explicit workflow pins before switching; the current supported-model guide<\/a> says Kimi K3 is excluded from default enablement, so verify its policy separately.<\/p>\n\n\n\n

September 23, 2026 update:<\/strong> GitHub’s September 18 notice<\/a> schedules six more Copilot models across Chat, inline edits, ask and agent modes, and code completions on October 19. The later date remains separate from the completed October 2 wave; its mappings and Business\/Enterprise default-enable rule are below.<\/p>\n\n\n\n

September 11, 2026 source date:<\/strong> GitHub’s VS Code Agents usage update<\/a> and code-review update<\/a> were published on September 11. Availability depends on plan, policy, repository and product surface.<\/p>\n\n\n\n

September 4, 2026 update:<\/strong> The Gemini 3.8 Flash<\/a> and Business and Enterprise signup<\/a> notices describe gradual rollout and seat or billing timing. Verify the current organization state before acting on either.<\/p>\n\n\n\n

September 3, 2026 update:<\/strong> Two September 2 Changelog entries add governance controls: enterprise-managed default models<\/a> and content exclusions in the Copilot app and CLI<\/a>. These controls cover different surfaces and policy scopes.<\/p>\n\n\n\n

What changed in the September 11 update<\/h2>\n\n\n\n

GitHub\u2019s usage-metrics entry adds fields for VS Code Agents, including daily active users and totals by the dedicated VS Code agent surface. GitHub distinguishes that surface from editor Agent Mode and generic Copilot rollups. A blank or null value can therefore mean that the feature is unavailable or that the field does not apply; it should not be silently converted into zero usage.<\/p>\n\n\n\n

A separate code-review entry describes addressed-comment auto-resolution, smart commit messages and analysis changes. GitHub also describes an agent ensemble in Lite and broader shell-tool work behind firewall controls. The percentages in the announcement are GitHub\u2019s own experiment results. Keep them labelled as vendor evidence, not as an independent benchmark or a promise that a repository will see the same outcome.<\/p>\n\n\n\n

A decision matrix for Copilot administrators<\/h2>\n\n\n\n
\n
Signal<\/th>What it can answer<\/th>What it cannot answer by itself<\/th><\/tr><\/thead>
VS Code Agents usage fields<\/td>Whether the dedicated surface is appearing in the selected usage view.<\/td>Whether another Copilot mode was unused, or whether the activity was valuable.<\/td><\/tr>\n
Addressed-comment automation<\/td>Which review comments a feature may attempt to resolve or explain.<\/td>Whether the code is safe, correct or approved for merge.<\/td><\/tr>\n
Weekly release entry<\/td>Which product surfaces deserve an entitlement and policy check.<\/td>Whether the feature is enabled for your plan, organisation or repository.<\/td><\/tr>\n<\/tbody><\/table><\/figure>\n\n\n\n

A safe admin checklist<\/h2>\n\n\n\n
    \n
  1. Identify the surface.<\/strong> Record whether the question concerns VS Code Agents, editor Agent Mode, the Copilot app, CLI, JetBrains or code review. Do not compare unlike counters.<\/li>\n
  2. Record the policy boundary.<\/strong> Note organisation, enterprise, repository and user settings that can control access. A changelog entry is not a live entitlement read-back.<\/li>\n
  3. Choose a review-only pilot.<\/strong> Select a low-risk repository, keep branch protection and required human review in force, and record the commit SHA and feature setting.<\/li>\n
  4. Inspect the proposed change.<\/strong> An addressed comment, smart commit message or analysis result still needs a human owner to validate scope, tests, security and attribution.<\/li>\n
  5. Measure the right denominator.<\/strong> Keep dedicated VS Code Agent fields separate from other Copilot usage. Document null handling and the time window before comparing reports.<\/li>\n
  6. Write a rollback rule.<\/strong> Disable the experiment or revert the workflow if review quality, policy compliance or repository safety is not acceptable.<\/li>\n<\/ol>\n\n\n\n

    What the weekly release does not prove<\/h2>\n\n\n\n

    The September 7 weekly release<\/a> lists Jira integration in the Copilot app, experimental HydraFusion CLI routing, VS Code automations, voice and issue work, and a managed JetBrains sandbox. A weekly roundup is a discovery pointer; it does not prove a shared availability date, identical permissions or production readiness across surfaces. Verify the current documentation and account state before changing a workflow.<\/p>\n\n\n\n

    September 2 additions: default models and content exclusions<\/h2>\n\n\n\n

    Default model policy<\/h3>\n\n\n\n

    GitHub\u2019s default-model changelog<\/a> says enterprise-managed settings support any preferred default model for new conversations, with team overrides through the documented model and team-mapping controls. The relevant administrator task is to record the intended default, the teams allowed to override it and the review owner. \u201cAvailable in the setting\u201d does not mean every model is available to every plan, organisation, user or surface. Check the current default-availability guide<\/a> and model-access guide<\/a> before applying the setting.<\/p>\n\n\n\n

    Content-exclusion policy<\/h3>\n\n\n\n

    GitHub also says content exclusions are generally available in the Copilot app and CLI and that the app\/CLI respect enterprise, organisation and repository policies. The Changelog entry<\/a> should be read with the content-exclusion documentation<\/a>: an administrator configures paths or fnmatch-style patterns, propagation can take up to 30 minutes, and the setting must be tested. The current documentation notes a limitation for Agent mode in IDEs, so do not claim that one exclusion configuration blocks every Copilot surface.<\/p>\n\n\n\n

    \n \n
    Control<\/th>Administrator evidence to save<\/th>Boundary<\/th><\/tr><\/thead>\n
    Default model<\/td>Enterprise setting, preferred model, team mappings and an example new conversation.<\/td>Plan, organisation and surface availability still need verification.<\/td><\/tr>\n
    Content exclusion<\/td>Enterprise\/org\/repository policy, path patterns, propagation timestamp and a controlled test.<\/td>App\/CLI GA does not establish full IDE Agent-mode coverage.<\/td><\/tr>\n
    Budget expiry<\/td>Owner, limit, expiry date, notification and renewal decision.<\/td>Model choice can affect consumption, but it does not change an individual budget limit, expiry or renewal policy.<\/td><\/tr>\n
    Code-review approval<\/td>Repository rule, reviewer role, pilot evidence and rollback path.<\/td>Model selection does not grant Copilot approval authority; repository rules and configured review settings determine whether an assessment can count toward required approvals.<\/td><\/tr>\n <\/tbody>\n<\/table><\/figure>\n\n\n\n

    Use GitHub\u2019s enterprise-managed-settings reference<\/a> as the schema authority and its configuration guide for the admin path. Test one repository with a known excluded path, one included path and the actual client surfaces your teams use. Save the observed policy state and timestamp; do not infer a successful exclusion from a setting that has not propagated. These controls have different scopes: model availability, excluded repository paths, budget expiry and code-review approvals each need a named owner and a current policy read-back.<\/p>\n\n\n\n

    Changelog posts and community discussions can help identify administrator questions, but they do not establish plan entitlement or enforcement. For adjacent implementation, use the Copilot VS Code release guide<\/a> for feature changes, the agent\/plugin guide<\/a> for tool governance and the agent-harness guide<\/a> for state and evaluation.<\/p>\n\n\n\n

    The two September 2 entries are documented product-policy updates. They do not prove that a particular organisation has the setting enabled, that a user can select every model or that an exclusion covers every client. Confirm account-level readback before changing production policy.<\/p>\n\n\n\n

    Short answer:<\/strong> GitHub Copilot retired four models on October 2 and has six more scheduled for October 19. Use the date-specific replacement map, check plan and model policy, then update any pinned workflows. Keep the September billing, budget, default-model, content-exclusion, usage and review controls in the same admin checklist.<\/p>\n\n\n\n

    Four GitHub Changelog entries from 31 August and 1 September 2026 change Copilot governance rather than adding one isolated feature. This checklist turns the model, access, budget and review-policy changes into explicit owner decisions.<\/p>\n\n\n\n

    The September admin checklist<\/h2>\n\n\n\n

    GitHub’s 31 August and 1 September 2026 Changelog entries form one practical administrator update. The changes touch four controls: selected model deprecations, Team-plan model access when a user belongs to multiple organisations, individual-user budget expiry, and Copilot code-review approvals.<\/p>\n\n\n\n

    The safest response is an inventory, not a panic migration. Check which repositories and teams depend on a named model, identify the billing organisation for multi-seat users, record budget owners and expiry dates, and decide whether code-review approval belongs in branch protection. GitHub says no action is required to remove the deprecated models, but that does not remove the need to test the replacement workflow.<\/p>\n\n\n\n

    This guide covers model availability and administrator controls. Use the GitHub Copilot VS Code August release guide<\/a> for editor features and the Fable versus Sonnet comparison<\/a> for a broader model-selection discussion.<\/p>\n\n\n\n

    Model lifecycle<\/th>Selected Copilot models are deprecated with replacement families listed by GitHub; no removal action is required.<\/td><\/tr>\n
    Team access<\/th>For users with multiple organisation seats, access is determined by the billing organisation under Usage billed to.<\/td><\/tr>\n
    Budget<\/th>Business and Enterprise admins can set an individual budget expiry or no expiry.<\/td><\/tr>\n
    Review<\/th>Copilot can make configurable approval assessments; approvals are off by default.<\/td><\/tr>\n
    Owner<\/th>The organisation should record a human owner for each policy and exception.<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

    Model deprecations: map by family, then test<\/h2>\n\n\n\n

    A date-aware model lifecycle register<\/h3>\n\n\n\n

    GitHub’s August 31 notice<\/a>, September 10 notice<\/a>, October 2 notice<\/a> and September 18 notice<\/a> give the dates and suggested alternatives below. The list is scoped to models inside GitHub Copilot; it does not announce retirement of the providers’ separate APIs. “Suggested alternative” is a migration pointer, not a promise that behavior, access or cost is equivalent.<\/p>\n\n\n\n

    The table preserves the alternatives GitHub named at each date; some later changed status. Claude Opus 4.7 was retired in the October 2 notice<\/a>, and Gemini 3.7 Flash is scheduled in the October 19 notice<\/a>. For current selection, follow the later dated rows and GitHub’s current supported-model documentation<\/a>.<\/p>\n\n\n

    \n\n
    Date and state<\/th>Copilot model<\/th>GitHub suggested alternative<\/th><\/tr><\/thead>\n
    September 1, 2026
    Retired in most Copilot experiences<\/td>    		Gemini 3.1 Pro<\/td>Gemini 3.7 Flash<\/td><\/tr>\n
    September 1, 2026
    Retired<\/td>    		Claude Opus 4.5<\/td>Claude Opus 4.7, Opus 4.8, or Opus 5<\/td><\/tr>\n
    September 1, 2026
    Retired<\/td>    		Claude Opus 4.6<\/td>Claude Opus 4.7, Opus 4.8, or Opus 5<\/td><\/tr>\n
    September 1, 2026
    Retired<\/td>    		Claude Sonnet 4.5<\/td>Claude Sonnet 5<\/td><\/tr>\n
    September 1, 2026
    Retired with an annual-plan exception<\/td>    		Claude Sonnet 4.6<\/td>Claude Sonnet 5<\/td><\/tr>\n
    September 1, 2026
    Retired<\/td>    		Raptor Mini<\/td>MAI-Code-1.1-Flash<\/td><\/tr>\n
    September 10, 2026
    Retired<\/td>    		MAI-Code-1-Flash<\/td>MAI-Code-1.1-Flash<\/td><\/tr>\n
    October 2, 2026
    Retired<\/td>    		Gemini 3.5 Flash<\/td>Gemini 3.8 Flash<\/td><\/tr>\n
    October 2, 2026
    Retired<\/td>    		Gemini 3.6 Flash<\/td>Gemini 3.8 Flash<\/td><\/tr>\n
    October 2, 2026
    Retired<\/td>    		Kimi K2.7 Code<\/td>Kimi K3<\/td><\/tr>\n
    October 2, 2026
    Retired<\/td>    		Claude Opus 4.7<\/td>Claude Opus 5.5<\/td><\/tr>\n
    October 19, 2026
    Scheduled<\/td>    		Gemini 3.7 Flash<\/td>Gemini 3.8 Flash<\/td><\/tr>\n
    October 19, 2026
    Scheduled<\/td>    		GPT-5.5<\/td>GPT-5.6 Sol<\/td><\/tr>\n
    October 19, 2026
    Scheduled<\/td>    		GPT-5.4<\/td>GPT-5.6 Sol<\/td><\/tr>\n
    October 19, 2026
    Scheduled<\/td>    		GPT-5.4 mini<\/td>GPT-5.6 Luna<\/td><\/tr>\n
    October 19, 2026
    Scheduled<\/td>    		GPT-5 mini<\/td>GPT-5.6 Luna<\/td><\/tr>\n
    October 19, 2026
    Scheduled<\/td>    		Grok 4.5<\/td>Grok 4.6<\/td><\/tr>\n<\/tbody><\/table><\/figure>\n\n\n\n

    September 1 and 10:<\/strong> GitHub noted that Claude Sonnet 4.6 remains available to individual Copilot subscribers on annual plans. Current supported-model documentation<\/a> retains that exception; check the current plan and client before assuming the old model is unavailable to every user.<\/p>\n\n\n\n

    October 2, effective:<\/strong> The effective-date notice<\/a> maps Claude Opus 4.7 to Opus 5.5. The September 3 advance notice<\/a> had suggested Opus 5, while the current supported-model retirement history<\/a> uses generic “latest Claude Opus” wording; use the October 2 notice for the version-specific mapping. GitHub says Enterprise administrators may need to enable alternatives. Its current supported-model guide excludes open-weight Kimi K3 from default enablement. Inspect Kimi K3’s explicit organization or enterprise model policy and confirm it appears in the user’s plan and client selector, using the model-access guide<\/a> for the setup path. For Opus 5.5 access and billing details, see the Copilot Opus 5.5 access and billing guide<\/a>.<\/p>\n\n\n\n

    October 19, scheduled:<\/strong> The October 19 notice<\/a> says the suggested alternatives are automatically enabled for Copilot Business and Enterprise under default model enablement unless an administrator turned off the global default or explicitly disabled a model. This condition belongs to the six October 19 models in the table; do not apply it to October 2’s Kimi K3 row. The default-availability guide<\/a> and model-access guide<\/a> explain current policy and plan boundaries. GitHub says Copilot Free and Student use auto model selection only.<\/p>\n\n\n\n

    Before changing a default:<\/strong> record the lifecycle date, current model, suggested target, owner, client surface and policy setting. Search repository settings, scripts, integrations and shared guidance for explicit Copilot selections, then review each match before editing so provider API identifiers are not changed by mistake. Check whether the target appears in the affected user’s model selector. Run representative tasks in a low-risk repository only after access is confirmed, then save the model, date, repository and observed result. No replacement-model or account-policy test is reported in this guide.<\/p>\n\n\n\n