> ## Documentation Index > Fetch the complete documentation index at: https://docs.apiyi.com/llms.txt > Use this file to discover all available pages before exploring further. # ChatHub > A browser-extension AI client that compares multiple model answers side-by-side; unified access to 400+ models via APIYI ChatHub is a Chrome / Edge browser extension that lets you ask multiple AI models the same question in one conversation and displays their answers side-by-side, so you can quickly judge which model best fits the current task. With APIYI, ChatHub can unify access to 400+ mainstream models including OpenAI, Claude, Gemini, DeepSeek, and Qwen. ## Core capability: side-by-side model comparison ChatHub's biggest differentiator from other clients is **multi-model parallel Q\&A**: * Send the same question to 2–6 models in one tab * View their answers in either **side-by-side** or **table** layout * Any combination works: GPT-5.5 + Claude Opus 4.7 + Gemini 3 Pro + DeepSeek V4 together * Comparison view supports Markdown rendering, code highlighting, and image previews **Best fit scenarios** * Writing code / fixing bugs: see GPT-5.5 and Claude Opus 4.7 patches at once * Long-form writing: compare Claude Sonnet and Qwen3.7 Chinese phrasing * Translation / summarization: feed the same paragraph to GPT-5.5, Gemini 3 Flash, and DeepSeek V4 * Model selection: stress-test multiple models with the same prompt to pick a default ## Install Visit `https://chathub.gg/zh-CN/download` to install ChatHub. ## Quick setup (OpenAI-compatible mode) This is the most universal setup and supports all 400+ models on APIYI. ### Step 1: Get an API key Refer to [API key creation and management](/en/faq/token-management) to get your APIYI key. ### Step 2: Open ChatHub settings 1. Click the ChatHub icon in the top-right corner of your browser 2. In the popup, click the **Settings** (gear) icon in the top-right 3. Go to the **Custom Providers** tab ### Step 3: Add the APIYI provider ChatHub add APIYI provider configuration example Following the screenshot above, fill in the add-provider dialog as follows: * **Name**: `apiyi` (anything that helps you tell this provider apart in the list) * **API Host**: pick `OpenAI` from the dropdown, and put `https://api.apiyi.com/v1` in the URL field below it * **API Key**: paste your APIYI key * **Model**: a model you use often, e.g. `deepseek-v4-pro` Optional advanced options: * **Image input**: enable if the model supports vision (e.g. `gpt-5-chat-latest`, `gemini-3-pro-preview`, `claude-sonnet-4-5`) * **Function calling**: enable if the model supports tool/function calls * **Advanced settings**: collapsed by default — expand to customize request headers, timeouts, etc. When done, click **Confirm** in the bottom-right to save. **Configuration notes** * The API Host URL field must include the `/v1` suffix * "Model" is the default; you can switch models anytime during a chat * No browser restart needed — settings take effect immediately * To add another provider, repeat this step ### Step 4: Pick a model After saving, return to the ChatHub main interface. The models you just added will appear in the dropdown — switch freely. ## Three usage modes | Mode | Trigger | When to use | | ----------------------- | -------------------------------------------------- | -------------------------------------------- | | **Single-model chat** | Talk to one model like a normal ChatGPT | Daily Q\&A, tasks one model excels at | | **Multi-model compare** | Tick 2–6 models on the right of the input box | Selection tests, finding the best answer | | **Web assistant** | Select text on any page to pop the ChatHub toolbar | Translate/summarize a paper, polish an email | **Tips for compare mode** * In compare mode each model's reply is **independent**; the left main panel still shows your default model * Want to export everyone's replies? Click the **Export** button in the top-right of the conversation — Markdown / JSON / PNG are all supported * Pick **stylistically distinct** models for comparison, e.g. "GPT-5.5 + Claude Opus 4.7 + DeepSeek V4" — differences are far easier to spot ## Advanced: the strongest compare setup ChatHub's killer feature is "run multiple vendors at once". Through APIYI you can mix any vendor's models in a single compare session. ### Recommended compare combinations | Task | Recommended mix | Why | | -------------------------------- | ------------------------------------------------ | --------------------------------------------- | | **Code generation / bug fixes** | GPT-5.5 + Claude Opus 4.7 + DeepSeek V4 | Top coding + a cost-effective domestic option | | **Long-form / creative writing** | Claude Sonnet 4.5 + Qwen3.7-Max + GPT-5.5 | Chinese expression + Western storytelling | | **Translation / multilingual** | GPT-5.5 + Gemini 3 Flash + DeepSeek V4 | Three vendors cross-validate | | **Reasoning / math** | GPT-5.5 (xhigh) + Claude Opus 4.7 + Gemini 3 Pro | Three top reasoning configs | ## Supported models Through APIYI, ChatHub supports 400+ mainstream AI models, including OpenAI, Google Gemini, Claude, DeepSeek, domestic models, and more. Latest model recommendations, performance comparisons, and scenario-based usage tips. Covers text creation, programming, fast response, image generation, video generation, and more. **Why no model list on this page?** Models iterate very quickly. To make sure you always get accurate recommendations, we maintain the latest list on the [Model recommendations page](/en/api-capabilities/model-info) along with benchmarks and usage tips. ## Core features ### Keyboard shortcuts | Shortcut | Action | | -------------------------- | ------------------------------------------ | | `Ctrl/Cmd + Shift + Y` | Open the ChatHub sidebar on any web page | | `Ctrl/Cmd + B` | Toggle sidebar visibility | | `Ctrl/Cmd + Enter` | Send message (single-model mode) | | `Ctrl/Cmd + Shift + Enter` | Send to all selected models (compare mode) | ### Web assistant Select any text on a web page and ChatHub pops a toolbar that can: * Translate the selected text * Summarize the selected paragraph * Rewrite / polish * Explain code ### Conversation management * Multiple independent conversations with their own contexts * Local storage of conversation history (no cloud dependency) * Markdown rendering and code highlighting * One-click export to Markdown / PDF ### Prompt library Built-in prompt template marketplace; you can also save your own reusable prompts and use variables. ## Advanced settings ### Parameter tuning ```yaml theme={null} Chat parameters: - Temperature: 0.7 # Creativity control (reasoning models like GPT-5 only support 1) - Max Tokens: 4096 # Maximum output length - Top P: 0.9 # Nucleus sampling threshold - Frequency Penalty: 0 # Repetition penalty - Presence Penalty: 0 # New-topic tendency ``` ### Streaming output Streaming is on by default in ChatHub for a smoother experience. ### Proxy settings If you need to access through a proxy: 1. Go to Settings > Advanced 2. Configure the HTTP/HTTPS proxy 3. Set proxy authentication if needed ## Troubleshooting ### Connection failed * Verify the API Base URL: `https://api.apiyi.com/v1` * Confirm the API key is valid * Check your network connection ### Models not showing * Wait for the model list to auto-refresh * Click the "Refresh models" button manually * Check API key permissions ### Compare mode shows only one response * Make sure you ticked multiple models on the right side of the input box * Confirm every model comes from a configured provider ### Slow response * Try a faster model * Check network latency * Shorten the context ## Best practices ### Performance optimization 1. **Pick models wisely** * Match model size to task complexity * See [Model recommendations](/en/api-capabilities/model-info) for the latest selection tips 2. **Compare strategy** * Comparing two similar models (two GPTs) tells you very little * Pick models with **different strengths**: one for coding, one for writing, one for reasoning, one for translation * Don't compare more than 4 models at once — token usage scales linearly 3. **Resource management** * Monitor API usage * Set spending alerts * Update the extension regularly ### Security tips * Never share API keys * Rotate keys periodically * Protect your browser with a strong password * Handle sensitive information carefully ### Team collaboration Although ChatHub is primarily for individuals, you can still support teams through: * Shared prompt templates * Exported conversation records * Standardized API configuration ## Comparison ### vs other clients | Feature | ChatHub | Single-model clients | Other extensions | | ------------------------------ | ----------------- | ------------------------ | ------------------------ | | **Side-by-side multi-model** | ✅ Core capability | ❌ | A few | | **Browser sidebar** | ✅ | ❌ | Some | | **Web assistant** | ✅ | ❌ | Some | | **Custom API** | ✅ | ✅ | ✅ | | **Cross-device sync** | ✅ (browser sync) | ❌ | Implementation dependent | | **Local conversation history** | ✅ | Implementation dependent | Implementation dependent | ### Why choose ChatHub 1. **Selection tool**: comparing multiple models at once is irreplaceable 2. **Multi-vendor mix**: combine OpenAI / Claude / Gemini / domestic models freely 3. **Web-native**: as a browser extension it's callable on any page 4. **Privacy first**: conversations stored locally; the extension doesn't collect your data 5. **Active development**: continuously maintained *** Need more help? Visit the ChatHub GitHub repository: `github.com/chathub-dev/chathub` or the APIYI website: `api.apiyi.com`.