Codex Web GPT v6.1.2

V6 brings a new model picker, optional Pro usage tracking, more control over browser chats, and the fixes developed since v5.0.8.

V6 highlights

• Clear model names and native Effort controls. Automatic models now end in (Web): GPT-5.6 Luna, GPT-5.6 Sol, GPT-5.6 Pro, and GPT-6 Pro. Sol exposes the reasoning levels your account supports through Codex's Effort selector. Instant remains a separate entry so its context budget is preserved. Existing tasks using the old model names continue to resolve to their original modes.
• A Limits tab for Pro. Opt in by checking your signed-in plan. The launcher tracks its own submitted messages for the supported Pro $100 and $200 plans, separates the two Pro models and shared allowances, and shows a yellow indicator from 75% usage. These are local estimates over rolling windows, not OpenAI's remaining quota or reset schedule; messages sent outside the launcher are not counted. Tracking is unavailable in Zero Risk.
• Smaller parts for Bigger Context. Large requests now use six ordered parts instead of three; smaller requests can still use one or two. The overall context and compaction budgets are preserved. The final part starts the task, and tools and attachments stay with that final part.
• New browser chat for each turn. An optional setting rebuilds the context of the same Codex task in a fresh ChatGPT conversation and attaches its connector again. This can help when ChatGPT stops retaining tools across follow-ups, at the cost of sending more context. Off by default; automatic mode only.
• Save chats in ChatGPT. An optional setting uses regular conversations instead of Temporary Chat. Off by default. ChatGPT memory and custom instructions may apply when enabled; it is independent of the fresh-chat setting.
• Native Linux ARM64 packages. Linux now has x64 and ARM64 AppImages and runtime bundles, with architecture-aware installation and updating.
• A new application icon. The launcher keeps its familiar ChatGPT mark inside the interface.

Fixes and improvements through v6.1.2

• ChatGPT replies and activity. Supports both ChatGPT layouts and keeps following replies when the page redraws. Fixes false “completed text changed” errors caused by empty blocks and code-block controls; code keeps its line breaks. Activity descriptions appear as grey status text in Codex, while actual messages stay separate.
• File attachments. Recognizes completed uploads so attachment checks do not leave Send waiting.
• Model and effort selection. Handles menus changing while they load, recognizes locked Pro options on Plus, and checks the requested effort before sending. Available Sol effort levels remain selectable when ChatGPT hides Pro.
• Sign-in and Limits. Detects signing out and switching accounts, including while the browser is hidden. When ChatGPT redirects a task to sign-in, the launcher now asks you to sign in instead of repeating navigation errors. Gives account controls time to load, opens the current Billing page, and recognizes personal ProLite and Pro 5x plans.
• Workspaces and long tasks. Improves workspace recovery and continuation. Accepts extra folders explicitly allowed by Codex, including repeated permission entries.
• Context summaries. Improves compaction and reports when ChatGPT finishes without submitting the required summary, instead of waiting for the full timeout. Clarifies the plugin instructions; ChatGPT-side safety refusals are not guaranteed resolved.
• Formatting. Preserves Markdown links, plan markers, and Windows file labels.
• Setup and updates. Improves Windows and tunnel installation, adds clearer Linux update instructions, and updates the plugin walkthrough with click-to-pause video and a pause indicator.
• Desktop behavior. Fixes the missing macOS menu-bar icon, restores windows after a monitor disconnects, preserves Activity across rotated logs, and improves handling of cancelled tabs.
• Diagnostics. Makes exports safer and preserves the original connection error when cleanup also fails. Model-loading logs distinguish cancellations and timeouts without exposing request details. A missing message field no longer automatically means an expired login.
• Plugin names. Change the part after Codex in Settings → Plugin name when ChatGPT will not let you reuse a deleted name. Automatic and Zero Risk names are saved separately. Create and verify a new plugin using the displayed name; your tunnel credentials and login are kept.
• Zero Risk recovery. Gives you 60 seconds to send. Copying the prompt again restarts that timer, and Copy remains available after Sent until the plugin starts.
• Local tool connections. Prevents failed startup cleanup from removing another running connection and rejects malformed replies explicitly.
• ChatGPT notices. Dismisses the observed announcement banner and clarifies how connected tools use Codex permissions.

Full-mode users: after updating, refresh your Codex plugin in ChatGPT's plugin settings to load the updated tool description.

Thanks to @renanassisalves for PR #664, page details, and testing that helped identify the ChatGPT interface changes. Thanks also to @NC-Channel-Coder and @eisen0419 for their proposed fixes, to @omarcosr and @Wandersport for their reports and proposals, and to everyone who shared logs, page details, and retested updates.

Install

Quit the launcher, download the package for your system, and replace the existing application. Reopen it and restart Codex after the launcher refreshes its integration. Do not delete your launcher data or ChatGPT profile.

• Windows x64
• macOS Apple silicon · macOS Intel
• Linux x64 · Linux ARM64