feat(proxy): add custom favicon support

[hztBUAA]

Summary

Adds support for configuring a custom favicon in the LiteLLM proxy admin UI, following the same pattern as the existing custom logo feature.

• Add favicon_url field to UIThemeConfig model
• Add LITELLM_FAVICON_URL environment variable support
• Add /get_favicon endpoint to serve custom favicons (with URL fetching and fallback to default)
• Update ThemeContext to dynamically set the browser favicon on page load
• Add favicon URL input to the UI Theme Settings page
• Add comprehensive backend tests (7 new tests)

Configuration

Via UI

Navigate to Settings > UI Theme, enter a favicon URL, and click Save.

Via Environment Variable

export LITELLM_FAVICON_URL=https://example.com/my-favicon.ico

Via Config YAML

litellm_settings:
  ui_theme_config:
    favicon_url: https://example.com/my-favicon.ico

Closes #8323

Test Plan

• test_get_ui_theme_settings_includes_favicon_schema - Verifies favicon_url appears in schema
• test_get_ui_theme_settings_with_favicon_configured - Verifies favicon_url returned when configured
• test_update_ui_theme_settings_with_favicon - Verifies setting favicon via PATCH
• test_update_ui_theme_settings_clear_favicon - Verifies clearing favicon via PATCH
• test_get_favicon_default - Verifies default favicon served
• test_get_favicon_with_custom_url - Verifies custom URL favicon fetching
• test_get_favicon_url_error_fallback - Verifies fallback on fetch error
• Existing test_get_ui_theme_settings and test_update_ui_theme_settings still pass

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Feb 20, 2026 5:40am

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[CLAassistant]

All committers have signed the CLA.

[greptile-apps]

Greptile Summary

Added custom favicon support to LiteLLM proxy admin UI, following the existing custom logo pattern. The implementation includes backend endpoint, configuration management, and frontend integration.

Key Changes:

• Added favicon_url field to UIThemeConfig model with environment variable support (LITELLM_FAVICON_URL)
• Created /get_favicon endpoint that fetches from custom URLs with fallback to default favicon
• Updated UI Theme Settings page to include favicon URL input alongside logo configuration
• Added ThemeContext integration to dynamically update browser favicon on page load
• Included 7 comprehensive tests (4 integration + 3 unit tests) with proper mocking

Recommendations:

• Consider adding caching for remote favicon fetches (similar to logo endpoint) to improve performance
• Resolve documentation inconsistency: field description says "Must be HTTP/HTTPS URL" but implementation also accepts local file paths

The PR successfully addresses issue #8323 with well-tested, maintainable code that integrates cleanly with existing theming infrastructure.

Confidence Score: 4/5

• This PR is safe to merge with minor performance optimizations recommended
• The implementation follows existing patterns (mirrors logo customization), includes comprehensive tests (7 new tests with proper mocking), and has no critical bugs. Score reduced by 1 for missing caching optimization that could impact performance at scale.
• litellm/proxy/proxy_server.py - consider adding caching for remote favicon fetches

Important Files Changed

Filename	Overview
litellm/proxy/proxy_server.py	Added /get_favicon endpoint to serve custom favicons with URL fetching and fallback logic, but missing caching optimization present in similar /get_image endpoint
litellm/proxy/ui_crud_endpoints/proxy_setting_endpoints.py	Added favicon_url field to UIThemeConfig model and update logic for LITELLM_FAVICON_URL environment variable, following existing pattern for logo configuration
ui/litellm-dashboard/src/contexts/ThemeContext.tsx	Added favicon state management and useEffect to dynamically update browser favicon when faviconUrl changes

Sequence Diagram

sequenceDiagram
    participant Browser
    participant ThemeContext
    participant UIThemeSettings
    participant ProxyAPI
    participant ProxyServer
    participant RemoteURL

    Browser->>ThemeContext: Load app
    ThemeContext->>ProxyAPI: GET /get/ui_theme_settings
    ProxyAPI-->>ThemeContext: {logo_url, favicon_url}
    ThemeContext->>ThemeContext: setFaviconUrl(url)
    ThemeContext->>Browser: Update <link rel="icon">

    Browser->>UIThemeSettings: Navigate to Settings
    UIThemeSettings->>ProxyAPI: GET /get/ui_theme_settings
    ProxyAPI-->>UIThemeSettings: Current theme config

    Browser->>UIThemeSettings: Enter favicon URL & Save
    UIThemeSettings->>ProxyAPI: PATCH /update/ui_theme_settings {favicon_url}
    ProxyAPI->>ProxyAPI: Update config & env vars
    ProxyAPI-->>UIThemeSettings: Success
    UIThemeSettings->>ThemeContext: setFaviconUrl(url)
    ThemeContext->>Browser: Update <link rel="icon">

    Browser->>ProxyServer: GET /get_favicon
    alt Custom URL configured
        ProxyServer->>RemoteURL: Fetch favicon
        RemoteURL-->>ProxyServer: Favicon content
        ProxyServer-->>Browser: Favicon (no caching)
    else No custom URL
        ProxyServer-->>Browser: Default favicon.ico
    end

Loading

_{Last reviewed commit: ba27242}

[greptile-apps]

_{6 files reviewed, 2 comments}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

Missing caching optimization that exists in get_image endpoint (lines 10694-10699). Every request fetches the favicon from the remote URL, which could cause performance issues. Consider adding file caching similar to logo handling.

_{Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!}

[greptile-apps]

Description states "Must be an HTTP/HTTPS URL" but get_favicon endpoint also accepts local file paths (proxy_server.py:10800-10807). Either update the description or add validation to enforce HTTP/HTTPS only.