Expose GPT-5.6 Max reasoning across ADE

[arul28]

Summary

• preserve Codex app-server Max reasoning tiers across desktop runtime catalogs
• expose Max in ADE Code and iOS, ordered before Ultra on Sol and Terra
• consolidate Model Picker helper coverage and document exact cross-surface ladders

Validation

• /quality: clean after one renderer catalog fix
• desktop affected tests: 595 passed
• desktop CI shards 1, 7, 8: 3,199 passed
• ADE Code TUI: 858 passed
• iOS focused simulator tests: 3 passed
• docs validation: 176 files

No release is included in this PR.

Summary by CodeRabbit

• New Features

  • Expanded GPT-5.6 model reasoning options to include Max and Ultra where supported.
  • Updated model pickers and catalogs to display the complete reasoning-effort range.
  • Improved model search, ordering, favorites, aliases, and runtime catalog handling.

• Bug Fixes

  • Corrected model-specific reasoning tiers and default selections across desktop, CLI, and iOS.

Greptile Summary

This PR exposes GPT-5.6 Max reasoning across ADE surfaces. The main changes are:

• Adds max to GPT-5.6 Sol, Terra, and Luna model metadata where supported.
• Preserves app-server advertised reasoning ladders in desktop runtime catalogs and ADE Code.
• Updates iOS Work and composer model catalogs to show the same GPT-5.6 tiers.
• Consolidates related model picker helper tests and updates cross-surface documentation.

Confidence Score: 5/5

Safe to merge with minimal risk.

Changes are narrowly scoped to model metadata, tier pass-through, display labels, tests, and documentation. Reviewed paths show consistent Max ordering and fallback behavior across desktop, ADE Code, and iOS.

No files require special attention.

T-Rex Logs

What T-Rex did

• The initial desktop Vitest run completed with passing test output as shown in the initial desktop log.
• To address the shell wrapper issue observed in the initial run, the desktop Vitest run was re-executed using bash -lc and completed successfully with EXIT_CODE 0.
• The ADE CLI based Vitest run completed successfully, as shown in the ADE CLI log with EXIT_CODE 0.

_{Ran code and verified through T-Rex}

Important Files Changed

Filename	Overview
apps/desktop/src/shared/modelRegistry.ts	Adds Max to static GPT-5.6 Codex reasoning tiers, before Ultra where supported.
apps/desktop/src/main/services/chat/agentChatService.ts	Stops filtering Max from Codex GPT-5.6 model info so advertised tiers pass through.
apps/desktop/src/renderer/components/shared/ModelPicker/modelCatalog.ts	Preserves runtime-advertised reasoning tiers including Max when building picker descriptors.
apps/ade-cli/src/tuiClient/modelState.ts	Exposes Max in Codex effort defaults and removes GPT-5.6-specific filtering while adding the Max label.
apps/ios/ADE/Views/Work/WorkModelCatalog.swift	Adds Max to curated iOS Work GPT-5.6 ladders and allows advertised Max defaults to pass through.
apps/ios/ADE/Views/Components/ADEDesignSystem.swift	Aligns iOS model reasoning tier lookup with desktop by adding GPT-5.6 Max.
apps/desktop/src/renderer/components/shared/ModelPicker/modelCatalog.test.ts	Consolidates model picker ordering/search/runtime-cache tests and adds coverage for complete GPT-5.6 ladders.
apps/ios/ADETests/ADETests.swift	Updates iOS focused tests for Max reasoning tiers and advertised Max defaults.

Sequence Diagram

%%{init: {'theme': 'neutral'}}%%
sequenceDiagram
participant AppServer as Runtime app-server catalog
participant Desktop as Desktop agentChatService
participant Registry as Shared modelRegistry
participant Picker as Desktop ModelPicker
participant TUI as ADE Code TUI
participant iOS as iOS Work catalog
AppServer->>Desktop: Advertise GPT-5.6 reasoningEfforts
Desktop->>Registry: Fallback to static GPT-5.6 descriptors
Desktop-->>Picker: Preserve low, medium, high, xhigh, max, ultra
Registry-->>TUI: Provide fallback reasoningTiers including max
Desktop-->>iOS: Send host catalog/defaultReasoningEffort
iOS->>iOS: Use advertised tiers or curated GPT-5.6 fallback including max

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
sequenceDiagram
participant AppServer as Runtime app-server catalog
participant Desktop as Desktop agentChatService
participant Registry as Shared modelRegistry
participant Picker as Desktop ModelPicker
participant TUI as ADE Code TUI
participant iOS as iOS Work catalog
AppServer->>Desktop: Advertise GPT-5.6 reasoningEfforts
Desktop->>Registry: Fallback to static GPT-5.6 descriptors
Desktop-->>Picker: Preserve low, medium, high, xhigh, max, ultra
Registry-->>TUI: Provide fallback reasoningTiers including max
Desktop-->>iOS: Send host catalog/defaultReasoningEffort
iOS->>iOS: Use advertised tiers or curated GPT-5.6 fallback including max

Loading

_{Reviews (2): Last reviewed commit: "ship: iteration 1 — update GPT-5.6 picke..." | Re-trigger Greptile}

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
ade	Ignored	Preview	Jul 10, 2026 3:38pm

[arul28]

@copilot review but do not make fixes

[coderabbitai]

📝 Walkthrough

Walkthrough

Changes

GPT-5.6 reasoning tiers

Layer / File(s)	Summary
Canonical GPT-5.6 tier catalogs apps/desktop/src/shared/modelRegistry.ts, apps/desktop/src/shared/modelRegistry.test.ts, apps/ios/ADE/Views/Components/ADEDesignSystem.swift, apps/ios/ADETests/ADETests.swift	Desktop and iOS model registries now expose updated Sol, Terra, and Luna reasoning ladders, including max and ultra where applicable.
Desktop and CLI effort propagation apps/desktop/src/main/services/chat/agentChatService.ts, apps/desktop/src/renderer/components/shared/ModelPicker/modelCatalog.ts, apps/ade-cli/src/tuiClient/modelState.ts, related tests	Advertised reasoning efforts pass through without GPT-5.6 Codex-specific max filtering, and CLI display labels include Max.
iOS catalog effort handling apps/ios/ADE/Views/Work/WorkModelCatalog.swift, apps/ios/ADETests/ADETests.swift	iOS advertised and fallback effort selection preserves max, including when used as a default reasoning effort.
CLI ladder and display expectations apps/ade-cli/src/tuiClient/__tests__/appInput.test.ts	CLI tests verify the distinct Sol, Terra, and Luna ladders and the updated reasoning setup-row text.

Model picker regression coverage

Layer / File(s)	Summary
Model picker behavior coverage apps/desktop/src/renderer/components/shared/ModelPicker/modelCatalog.test.ts	Adds coverage for GPT-5.6 ladders, model ordering, favorites, search matching and ranking, aliases, and SDK/CLI runtime catalog freshness.

Estimated code review effort: 3 (Moderate) | ~25 minutes

Possibly related PRs

• arul28/ADE#760: Updates the same GPT-5.6 Codex reasoning-tier visibility and max handling paths.

Suggested labels:desktop, ios

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately captures the main change: exposing GPT-5.6 Max reasoning across ADE surfaces.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/gpt-5-6-max-reasoning

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[arul28]

@codex review

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Breezy!

Reviewed commit:be1303fd1b

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".