Skip to content

Add chatgpt-web provider: route ChatGPT Web models via codex-chatgpt-web bridge #876

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
GENEXIS-AI wants to merge 1 commit into from
Closed

Add chatgpt-web provider: route ChatGPT Web models via codex-chatgpt-web bridge #876

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 29 additions & 0 deletions src/providers/registry.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1282,6 +1282,35 @@ export const PROVIDER_REGISTRY: readonly ProviderRegistryEntry[] = [
// A self-hosted proxy may legitimately run without a master key.
keyOptional: true,
},
{
id: "chatgpt-web",

@chatgpt-codex-connector chatgpt-codex-connector Bot Aug 2, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Retarget this provider change to dev

The reviewed commit is based directly on 88589c9 (release: v2.9.1) rather than the integration line, and its own subject marks it [WRONG BRANCH]. Rebase or recreate this change on dev before opening or merging it so the feature does not bypass the repository's single integration branch and current development changes.

AGENTS.md reference: AGENTS.md:L177-L178

Useful? React with 👍 / 👎.

label: "ChatGPT Web (codex-chatgpt-web)",
baseUrl: "http://127.0.0.1:17841/v1",
adapter: "openai-responses",
authKind: "local",

@chatgpt-codex-connector chatgpt-codex-connector Bot Aug 2, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Expose the local provider in dashboard presets

When a user opens the dashboard's Add Provider catalog, this entry is absent because deriveProviderPresets() includes only featured, key-auth, or dashboardPreset entries, while this provider is local and sets neither flag. Add dashboardPreset: true (or intentionally make it featured) so the new provider can actually be selected from the dashboard.

AGENTS.md reference: src/AGENTS.md:L18-L18

Useful? React with 👍 / 👎.

allowPrivateNetworkByDefault: true,
allowBaseUrlOverride: true,
keyOptional: true,
Comment on lines +1292 to +1293

@chatgpt-codex-connector chatgpt-codex-connector Bot Aug 2, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Update the registry parity assertions for these flags

With this entry present, tests/provider-registry-parity.test.ts deterministically fails: its optional-key assertion at line 444 does not include chatgpt-web, and its base-URL-override assertion at line 487 does not include it either. Remove the redundant keyOptional flag if local auth already expresses the intended behavior, and update the override invariant plus focused provider coverage so bun run test remains green.

AGENTS.md reference: src/AGENTS.md:L24-L26

Useful? React with 👍 / 👎.

liveModels: true,
models: ["chatgpt-web/light", "chatgpt-web/medium", "chatgpt-web/high", "chatgpt-web/extra-high", "chatgpt-web/pro"],
defaultModel: "chatgpt-web/high",
modelContextWindows: {
"chatgpt-web/light": 256_000,
"chatgpt-web/medium": 256_000,
"chatgpt-web/high": 256_000,
"chatgpt-web/extra-high": 256_000,
"chatgpt-web/pro": 256_000,
},
modelInputModalities: {
"chatgpt-web/light": ["text", "image"],
"chatgpt-web/medium": ["text", "image"],
"chatgpt-web/high": ["text", "image"],
"chatgpt-web/extra-high": ["text", "image"],
"chatgpt-web/pro": ["text", "image"],
},
dashboardUrl: "https://github.com/miuuyy/codex-chatgpt-web",

@chatgpt-codex-connector chatgpt-codex-connector Bot Aug 2, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Document how to install and configure the bridge

This provider requires a separately installed loopback bridge, but the repository-wide docs search finds no chatgpt-web setup instructions, and docs-site/src/content/docs/guides/providers.md still lists only Ollama, vLLM, and LM Studio under local providers. Add the bridge installation/startup steps, the supported CLI setup command, and the default endpoint to the English provider guide, keeping translated guides from contradicting it.

AGENTS.md reference: AGENTS.md:L192-L193

Useful? React with 👍 / 👎.

note: "Local ChatGPT Web browser bridge. Requires codex-chatgpt-web running on loopback (default port 17841). Routes ChatGPT Web Instant–Pro through an embedded browser session; no API key needed.",
},
{
id: "ollama-cloud",
label: "Ollama Cloud",
Expand Down
Loading