
Build a Custom AI Platform: A Complete Illustrated Guide to Integrating B.AI Models into WorkBuddy

In today's fast-paced development environment, WorkBuddy has become an indispensable "productivity hub" for many developers on the desktop due to its flexible workflow orchestration and powerful system-level integration capabilities. It not only aggregates fragmented development tools but also serves as a dedicated intelligent middleware, enabling developers to directly access top-tier AI capabilities within an immersive, context-switch-free environment. This significantly eliminates friction in executing daily tasks, allowing core focus to return entirely to high-value thinking and creation.
To help developers more efficiently integrate B.AI's high-performance model matrix and system-level infrastructure into their daily development workflows, this article will provide a detailed guide on how to add a custom B.AI model to the WorkBuddy desktop client. Before configuration, please prepare your WorkBuddy account, B.AI API Key, and the actual available model ID for your current account. Simply follow the straightforward steps below to unlock an exceptionally smooth AI collaboration experience locally.
I. Before You Begin
Step 1: Install WorkBuddy
Visit the WorkBuddy Official Download Page and select the appropriate installer based on your system environment. Users who already have WorkBuddy installed can skip this step. If you cannot find the "Models" or "Add Model" option, first select "Check for Updates" to upgrade to the latest version.
Windows
The official page currently provides the Windows x64 (compatible with ARM64) installer, requiring Windows 10 or later. After downloading the installer, double-click it and follow the wizard to complete the installation, then launch WorkBuddy.
Note: If the system blocks the installation, first verify that the installer came from the official page, then check the application name and publisher information in the popup dialog. Do not bypass checks by disabling Windows security protections.
macOS
The official page currently provides separate .dmg installers for Apple Silicon and Intel, requiring macOS 12.0 or later.
- Select the Apple Silicon version for M1, M2, M3, M4, and similar models.
- Select the Intel version for Intel processor models.
After opening the .dmg file, drag WorkBuddy into the "Applications" folder, then launch it from there. If you are unsure of your chip type, check the "Chip" or "Processor" section in "About This Mac".
Notes on Other Systems
Currently, the WorkBuddy desktop client only supports Windows and macOS and does not support Linux. Additionally, given the differing feature sets across mobile and HarmonyOS platforms, all operations in this tutorial apply exclusively to the desktop client.
Step 2: Log in to WorkBuddy
On the first launch, click "Login" and complete authentication using the method shown on the client. The international official documentation lists Google and GitHub OAuth. If the current client displays WeChat QR code scanning or other entry points, use the options actually presented by your client.
Step 3: Obtain Your B.AI API Key
Log in to B.AI first, navigate to the API Management or API Key Management page via the left sidebar, click "Create API Key", and assign it a recognizable name (e.g., WorkBuddy). Due to B.AI's official mechanism limitations, the full Key is displayed only once upon successful creation. Be sure to copy it immediately and store it securely.

Important confirmations before configuration:
- Prerequisites: Before configuration, verify that your account has sufficient balance/credits and that this API Key has permission to call the target model.
- Model ID Retrieval: The model ID must be sourced from the response of B.AI's GET /v1/models or the current console list, rather than relying solely on example names from other tutorials.
- Protocol Compatibility: WorkBuddy custom models utilize the OpenAI Chat Completions protocol, so the target model must also expose this endpoint. Configurations supporting only Anthropic Messages or OpenAI Responses cannot be directly applied.
Note: Never expose the full Key in articles, screenshots, chat records, or public repositories. If you suspect the Key has been compromised, immediately delete the old Key, create a new one, and update the WorkBuddy configuration.
II. Configure B.AI API via Interface
Step 1: Open Custom Model Configuration
Open WorkBuddy, click the account avatar in the bottom-left corner, and select "Settings".

Select "Models" on the left sidebar and click "Add Model".

If other custom models are already in the list, ensure you click "Add Model" to create a new configuration and do not overwrite parameters of unrelated models. To modify an existing B.AI model, simply click the pencil icon next to it. If you wish to keep the previous model while making changes, treat it as a brand-new model and add a separate configuration.
Step 2: Fill in Connection Parameters
Select "Custom" from the "Provider" dropdown menu and fill in the relevant information on the pop-up page.

Filling Instructions:

URL and "Custom Protocol" Must Be Set Pairwise:

According to the WorkBuddy official documentation, when "Custom Protocol" is disabled, the client automatically appends the `/chat/completions` path to the entered URL following standard OpenAI Chat Completions rules; when this toggle is enabled, the client sends requests directly to the complete URL you provide without any path concatenation. Do not mix these two methods. Incorrectly concatenating them into `/v1/chat/completions/chat/completions` will typically result in a 404 error.
Step 3: Configure Model Capabilities
Toggles under "Advanced Settings" are used solely to declare model capabilities. Enabling them will not grant unsupported models automatic functionality. Configure according to actual conditions, referencing the following principles:

When unable to confirm input/output limits, retain the "Use Provider Defaults" option. Without confirmation from documentation or actual testing, do not enable all capabilities by default.
Step 4: Save Configuration
After verifying the URL, API Key, and model ID, click "Save". The new model should appear in the "Saved Models" list.

If it does not display, sequentially check whether the save dialog is still open, if any fields show errors, if the model selector has refreshed, and verify the client version. If necessary, fully quit and restart WorkBuddy. The appearance of a model in the list only confirms that the configuration was saved; it does not prove that the B.AI API call succeeded, so subsequent validation is required.
III. Select Model and Verify Configuration
Step 1: Select the B.AI Custom Model
Return to "New Task", open the model selector near the input box, and choose the newly added model from the custom models group. During testing, do not select "Auto" mode, as it may route requests to other models and fails to prove that this request utilized B.AI.

Step 2: Verify Standard Conversation
Send a simple question that does not rely on tools:
Reply only with: B.AI standard conversation test successful
Upon receiving a normal response, it indicates that WorkBuddy has successfully read the configuration and that the Key, URL, and model ID can at least complete one text generation call.
If it fails, check in the order of "Current Model → API Key → URL & Protocol Toggle → Model ID → Balance & Permissions". Do not judge whether routing succeeded by asking the model "Who are you", as self-reported identity cannot serve as evidence of successful integration.
Step 3: Verify Tool Usage
First, create a new folder dedicated solely to testing, place one or two text files containing no private information inside, then select this folder as the `workspace` in WorkBuddy, granting only the permissions necessary to complete the reading task. Finally, send the following prompt:
Please read the text files in the current workspace, list the filenames, and summarize each file's content in one sentence. Do not modify, move, or delete any files.
Only when WorkBuddy displays a file-reading tool usage invocation and correctly returns the file summary does it indicate that the tool usage chain is operational. If it can chat but cannot read files, check tool usage, the model's tool-calling capability, B.AI endpoint support, and workspace permissions.
Step 4: Cross-check B.AI Call Logs
If the B.AI console provides usage or call logs, visit it to cross-check recent request timestamps, model IDs, request counts, and token consumption to ensure the data matches the just-completed test.
To ensure the entire integration process is fully functional, please verify the success status of the following three stages individually:
IV. Frequently Asked Questions
1. What to do if the custom model entry cannot be found?
This may be due to being unlogged, an outdated client version, or accessing non-desktop model settings. Log in first, then navigate to "Settings" → "Models". If the entry is still missing, select "Check for Updates" to update the client and restart. If the issue persists, submit the version number and screenshot via "Help & Feedback".
2. What to do if Windows or macOS installation/launch fails?
First verify that your system version meets requirements, the installer is from the official page, and the downloaded architecture is correct. Handle blocking prompts according to the system's built-in security procedures. Do not disable security protections or switch to unknown mirrors.
3. Why doesn't the model appear after saving?
First confirm whether the entry exists in "Saved Models", then close settings and reopen the model selector. If it still does not appear, fully quit and restart WorkBuddy, and check for client updates simultaneously.
4. What causes 401, 403, 404, or 429 errors, and how to resolve them?

Modify only one variable at a time, then retest with a standard conversation to facilitate root cause identification.
5. How to resolve the "model not found" prompt?
Rerun `GET /v1/models` or view the current console list to copy the exact `id`. Do not enter display names, aliases, or version numbers mentioned in other tutorials. Also confirm that the model is exposed to the Chat Completions endpoint.
6. What to do if it keeps loading, times out, or fails to connect?
First verify that your network can reach `https://api.b.ai`, then test with a brief prompt. Subsequently check the URL, proxy or corporate network policies, B.AI service status, and whether input files are too large. If the issue persists, record the timestamp, WorkBuddy version, complete error message, and request ID before contacting official support.
7. What to do if it can chat but cannot read files?
Check in the order of "Model Capability → Tool Calling Toggle → Workspace → File Permissions". Test with standard text files first; do not directly select system directories, protected directories, or sensitive files.
8. Why is the system still calling built-in models?
Please check the model selector to ensure it is not in "Auto" mode and that the configured B.AI custom model is explicitly selected. Send a short prompt, then visit the B.AI console to cross-check call timestamps and usage data. If doubts remain, it is recommended to create a new conversation task to retest.
9. How to launch it next time after closing the software?
On Windows, launch from the Start Menu or desktop shortcut; on macOS, launch from Applications or Launchpad. After restarting, if the custom model is not visible, first confirm whether the current account is logged in, then navigate to "Settings" → "Models" to check the saved models list.
10. Does switching computers require reconfiguration?
Reconfiguration is recommended. The WorkBuddy interface indicates that custom models are written to the local `.workbuddy/models.json` file. Do not assume API Keys and model configurations will sync merely by logging into the same account. On the new computer, you should re-add the model and complete both verification steps. Do not transmit Keys via chat, public cloud drives, or unencrypted documents.
References:
- WorkBuddy Official Download Page
- WorkBuddy Official Model Configuration Documentation
- WorkBuddy Windows Installation Guide
- WorkBuddy macOS Installation Guide
- B.AI API Reference
- B.AI Guide to Obtaining API Keys
















