codex-app-transfer
Local desktop gateway that translates OpenAI Codex CLI's Responses API into Chat Completions for Kimi / DeepSeek / Zhipu GLM / Bailian and other OpenAI-compatib
What is codex-app-transfer?
Local desktop gateway that translates OpenAI Codex CLI's Responses API into Chat Completions for Kimi / DeepSeek / Zhipu GLM / Bailian and other OpenAI-compatible providers.
Overview of codex-app-transfer
What it is
Codex App Transfer is a local desktop gateway and configuration tool for OpenAI Codex APP. It acts as a transparent proxy running on the local machine (127.0.0.1:18080) that translates the Responses API requests (HTTP streaming/non-streaming + /responses) from Codex APP into Chat Completions and other formats, forwarding them to a user-selected provider. It provides a desktop UI to manage providers, model mappings, forwarding ports, and logs, enabling Codex APP to seamlessly use third-party inference services that support the chat/completions protocol. When closing the window, it minimizes to the system tray and continues running; right-clicking the tray icon and selecting 'Exit' fully terminates the application. The current version is v2.4.5.
Key features
- Translates OpenAI Codex CLI's Responses API into Chat Completions for providers including Kimi, DeepSeek, Zhipu GLM, Bailian, and other OpenAI-compatible providers.
- Supports multiple protocol conversions: Chat Completions, Gemini Native (:streamGenerateContent), Gemini CLI OAuth (Cloud Code Assist), Anthropic Messages (/v1/messages), Grok Web, and Responses passthrough.
- Manages multiple providers and maps OpenAI model names (e.g., gpt-5.5, gpt-5.4) to actual provider model IDs via a desktop UI.
- Handles multi-turn tool conversations, previous_response_id history replay, autocompact expansion, and thinking/reasoning_content injection aligned with the OpenAI Responses API protocol.
- Includes optional features for Codex Desktop: background themes (11 anime themes + custom upload), usage display (quota/balance/tokens/cache), and a draft stash feature (stash/restore text+images).
Use cases
- Connecting OpenAI Codex APP to third-party LLM inference services that only support chat/completions or other non-Responses API protocols.
- Testing or using Codex with providers such as Kimi, DeepSeek, Zhipu GLM, Bailian, MiniMax, and others listed in the configuration.
- Developers who need a local gateway to bridge Codex APP with custom or self-hosted model endpoints.
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
This project is licensed under the MIT license. Pricing is open source.
Based on the project's public README and repository metadata.