docs: 添加硅基流动配置指南的中英文文档

Author: codedogQBY

website/src/content/docs/support/ai/siliconflow.md

@@ -0,0 +1,233 @@
+---
+draft: false
+title: SiliconFlow Configuration Guide
+description: Configure AI chat and remote embedding models with SiliconFlow.
+---
+
+[SiliconFlow](https://siliconflow.cn/) is a leading AI model service platform in China, providing rich open-source model APIs, including chat models and Embedding models.
+
+## Why Choose SiliconFlow?
+
+- **Affordable Pricing**: Pay-as-you-go, with free credits for new users
+- **Rich Model Selection**: Supports DeepSeek, Qwen, Llama and other mainstream open-source models
+- **Fast Access in China**: Servers located in China for fast access
+- **Embedding Models**: Provides various Embedding models for semantic search
+
+## Register Account
+
+1. Visit [SiliconFlow Official Website](https://cloud.siliconflow.cn/)
+2. Click "Register" to create an account
+3. Complete identity verification (required for some models)
+
+![SiliconFlow Register Page](./images/siliconflow-register.png)
+*Register SiliconFlow Account*
+
+![Identity Verification Page](./images/siliconflow-verify.png)
+*Complete Identity Verification*
+
+## Get API Key
+
+1. After logging in, go to [API Key Management](https://cloud.siliconflow.cn/account/ak)
+2. Click "Create API Key"
+3. Copy the generated API Key (starts with `sk-`)
+
+![API Key Management Page](./images/siliconflow-api-keys.png)
+*API Key Management Page*
+
+![Create New Key](./images/siliconflow-create-key.png)
+*Click Create API Key*
+
+![Copy API Key](./images/siliconflow-copy-key.png)
+*Copy Generated API Key*
+
+:::warning
+Please keep your API Key safe and do not share it with others.
+:::
+
+## Configure AI Chat
+
+### Steps
+
+1. Go to **Settings → AI**
+2. Click "Add Endpoint"
+3. Select **SiliconFlow** as provider
+4. Enter your API Key
+5. Click "Fetch Models" to get available models
+6. Select the model you want to use
+
+![ReadAny AI Settings Page](./images/readany-ai-settings.png)
+*ReadAny AI Settings Page*
+
+![Add SiliconFlow Endpoint](./images/readany-add-siliconflow.png)
+*Add SiliconFlow Endpoint*
+
+![Enter API Key](./images/readany-enter-apikey.png)
+*Enter API Key and Fetch Models*
+
+![Select Model](./images/readany-select-model.png)
+*Select Model to Use*
+
+### Recommended Models
+
+| Model | Description |
+|---|---|
+| **DeepSeek-V3** | Strong overall capability, suitable for daily chat |
+| **DeepSeek-R1** | Strong reasoning ability, suitable for complex analysis |
+| **Qwen2.5-72B-Instruct** | Alibaba open-source, strong Chinese capability |
+| **Llama-3.3-70B-Instruct** | Meta open-source, strong English capability |
+
+## Configure Embedding Model (Semantic Search)
+
+SiliconFlow provides various Embedding models for book semantic search functionality.
+
+### Steps
+
+1. Go to **Settings → Embedding Model**
+2. Select "Remote API"
+3. Select **SiliconFlow** as provider
+4. Enter your API Key
+5. Select an Embedding model
+
+![Embedding Model Settings Page](./images/readany-embedding-settings.png)
+*Embedding Model Settings Page*
+
+![Select SiliconFlow Embedding Model](./images/readany-embedding-siliconflow.png)
+*Select SiliconFlow as Embedding Model Provider*
+
+### Recommended Embedding Models
+
+| Model | Max Length | Description |
+|---|---|---|
+| **BAAI/bge-m3** | 8192 | Multilingual support, recommended for Chinese users |
+| **BAAI/bge-large-zh-v1.5** | 512 | Chinese-specific, fast |
+| **BAAI/bge-large-en-v1.5** | 512 | English-specific |
+| **Qwen/Qwen3-Embedding** | 32768 | Alibaba open-source, ultra-long text support |
+
+### Using Embedding Model
+
+After configuration:
+
+1. Open a book
+2. Click the "Vectorize" button in the sidebar
+3. Select "Use Remote API"
+4. Wait for processing to complete
+
+![Book Vectorize Button](./images/readany-vectorize-button.png)
+*Click Vectorize Button*
+
+![Select Remote API](./images/readany-vectorize-remote.png)
+*Select Use Remote API for Vectorization*
+
+![Vectorization Progress](./images/readany-vectorize-progress.png)
+*Vectorization Processing Progress*
+
+Embedding data will be sent to SiliconFlow servers for processing, and the index will be saved locally after completion.
+
+:::info
+Mobile only supports remote embedding API, not local embedding models (to reduce app size).
+:::
+
+## Pricing
+
+SiliconFlow uses pay-as-you-go pricing:
+
+- **Chat Models**: Charged by input/output token count
+- **Embedding Models**: Charged by processed token count
+
+New users get free credits after registration. Check the official website for details.
+
+![SiliconFlow Billing Page](./images/siliconflow-billing.png)
+*Billing and Usage Statistics Page*
+
+## Troubleshooting
+
+### API Key Invalid
+
+**Symptoms**: "API Key invalid" or "Authentication failed" error
+
+**Solutions**:
+- Confirm the API Key is correctly copied without extra spaces
+- Check if identity verification is completed
+- Confirm account balance is sufficient
+- Try creating a new API Key
+
+### Model List Empty
+
+**Symptoms**: Model list is empty or errors after clicking "Fetch Models"
+
+**Solutions**:
+- Check network connection
+- Confirm API Key is valid
+- Some models need to be enabled in SiliconFlow console first
+- Try manually entering the model name
+
+### Vectorization Failed
+
+**Symptoms**: Vectorization process interrupted or errors
+
+**Solutions**:
+- Check if embedding model is correctly configured
+- Confirm account balance is sufficient
+- Large files may take longer to process, please wait patiently
+- Check network connection stability
+- Try switching to another Embedding model
+
+### Connection Timeout
+
+**Symptoms**: Request timeout or no response
+
+**Solutions**:
+- Check network connection
+- If overseas, may need to use a proxy
+- Check if SiliconFlow service is normal (check official announcements)
+- Try changing network environment
+
+### Model Not Responding
+
+**Symptoms**: No reply for a long time after sending message
+
+**Solutions**:
+- Check if model is available in current region
+- Confirm model name is correct
+- Try switching to another model
+- Check if account has sufficient balance
+
+### Insufficient Balance
+
+**Symptoms**: "Insufficient balance" or "Quota exhausted" error
+
+**Solutions**:
+- Login to SiliconFlow console to check balance
+- Top up account
+- Check if there are unclaimed free credits
+
+### Rate Limit
+
+**Symptoms**: "Too many requests" or "Rate Limit" error
+
+**Solutions**:
+- Reduce request frequency
+- Wait for a while and retry
+- Upgrade account level for higher limits
+
+## Screenshot Checklist
+
+Please prepare the following screenshots and place them in `website/src/content/docs/support/ai/images/` directory:
+
+| Filename | Description |
+|---|---|
+| `siliconflow-register.png` | SiliconFlow register page |
+| `siliconflow-verify.png` | Identity verification page |
+| `siliconflow-api-keys.png` | API key management page |
+| `siliconflow-create-key.png` | Create new key popup |
+| `siliconflow-copy-key.png` | Copy API key |
+| `readany-ai-settings.png` | ReadAny AI settings page |
+| `readany-add-siliconflow.png` | Add SiliconFlow endpoint |
+| `readany-enter-apikey.png` | Enter API key |
+| `readany-select-model.png` | Select model |
+| `readany-embedding-settings.png` | Embedding model settings page |
+| `readany-embedding-siliconflow.png` | Select SiliconFlow embedding model |
+| `readany-vectorize-button.png` | Book vectorize button |
+| `readany-vectorize-remote.png` | Select remote API |
+| `readany-vectorize-progress.png` | Vectorization progress |
+| `siliconflow-billing.png` | Billing statistics page |