### Install Ollama on Linux

Source: https://docs.cherry-ai.com/pre-basic/providers/ollama

This command downloads and executes the Ollama installation script directly from the official website, providing a quick and convenient way to set up Ollama on Linux systems.

```sh
curl -fsSL https://ollama.com/install.sh | sh
```

--------------------------------

### Download and Run an Ollama Model

Source: https://docs.cherry-ai.com/pre-basic/providers/ollama

This command downloads and automatically runs a specified large language model, such as 'llama3.2', using Ollama. If the model is not present locally, Ollama will download it first, then start running it.

```sh
ollama run llama3.2
```

--------------------------------

### Viewing CherryStudio Console Errors for Network Requests

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/questions-and-feedback/faq

This guide details how to access and navigate the developer console in CherryStudio to diagnose network-related errors. It covers the keyboard shortcut to open the console, steps to inspect 'Network' requests like 'completions' or 'generations' for errors, and how to view the full response content for debugging purposes. It also lists specific network request names for different scenarios.

```User Interface Steps
1. Open Console: Ctrl+Shift+I (Mac: Command+Option+I)
2. Navigate: Network tab -> Click last "completions" or "generations" (marked with red ×)
3. View Details: Click "Response" tab to see complete return content.

Network Request Names by Scenario:
- Dialog, translation, model checking: completions
- Painting: generations
- Knowledge base creation: embeddings
```

--------------------------------

### Troubleshooting Knowledge Base Creation and Embedding Failures

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/questions-and-feedback/faq

This section outlines common reasons for failures in creating knowledge bases or obtaining embedding dimensions, primarily focusing on incorrect model usage. It clarifies that different model types (embedding, dialog, drawing) have distinct functions and should not be interchanged. It also provides guidance on how to correctly categorize embedding models within CherryStudio's model list if automatic categorization fails, and advises checking the service provider's documentation for model capabilities.

```Troubleshooting Points
Common Reasons for Failure:
1. Model state unavailable
2. Non-embedding model is used

Important Considerations:
1. Embedding models, dialog models, drawing models, etc. have their own functions, and their request methods and return contents and structures are different, so please don't force other types of models to be used as embedded models;
2. CherryStudio will automatically categorize embedding models. If a confirmed embedding model is not categorized correctly, go to the model list, click the Settings button for the corresponding model, and check the embedding option;
3. If you cannot confirm which models are embedding models, check the model information from the corresponding service provider.
```

--------------------------------

### Resolving Image Recognition and Upload Issues with Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/questions-and-feedback/faq

This section addresses problems where models fail to recognize images or users are unable to upload/select images. It emphasizes the necessity of confirming whether a model supports image recognition, identifiable by a small eye icon in CherryStudio. It also provides instructions on how to manually enable the image option for a model in its settings if its function is not correctly matched, and advises consulting the service provider's documentation for model-specific capabilities.

```Troubleshooting Points
1. Confirm Model Support: Check if the model supports image recognition (indicated by a small eye icon).
2. Correct Model Categorization: If the model function is not correctly matched, find the model in the model list, click the settings button, and check the 'image' option.
3. Consult Provider Documentation: For specific model information, refer to the corresponding service provider's documentation. Models that do not support vision do not need to enable the image function.
```

--------------------------------

### List Downloaded Ollama Models

Source: https://docs.cherry-ai.com/pre-basic/providers/ollama

This command displays a list of all large language models that have been downloaded and are available locally through Ollama, along with their sizes and other relevant details.

```sh
ollama list
```

--------------------------------

### Configure Assistant Prompt

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

This setting defines the prompt for the assistant. Users can refer to prompt writing guidelines on the agents page to effectively edit the content, influencing the assistant's behavior and responses.

```APIDOC
`Prompt`: That is, the prompt. You can refer to the prompt writing on the agents page to edit the content.
```

--------------------------------

### Configure Default Model for Assistant

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

Defines how a default model is assigned to an assistant, prioritizing the assistant's specific setting over a global default. This model is used when adding or copying an assistant.

```APIDOC
Parameter: Default Model
  description: Fixes a default model for an assistant.
  priority: Assistant's default model > Global Default Dialogue Model
  usage:
    - When adding from agent page.
    - When copying an assistant.
  fallback: If not set, initial model defaults to Global Default Dialogue Model.
```

--------------------------------

### Configure Context Window Size

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

Defines the number of messages retained in the model's context, directly impacting memory and token consumption.

```APIDOC
Parameter: Context Window
  description: Number of messages to keep in the context.
  impact: Larger value increases context length and token consumption.
  suggested_values:
    - 5-10: Suitable for normal conversations.
    - >10: Suitable for complex tasks requiring longer memory (e.g., generating long articles step-by-step with coherent logic).
  note: More messages consume more tokens.
```

--------------------------------

### Configure Multiple API Keys for Polling

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/settings/providers

This snippet shows the format for entering multiple API keys in CherryStudio, separated by English commas, to enable cyclic polling usage.

```plaintext
sk-xxxx1,sk-xxxx2,sk-xxxx3,sk-xxxx4
```

--------------------------------

### Enable Quick Translation via Spacebar Taps

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

This feature allows users to quickly translate entered content into English by tapping the spacebar three times after typing a message. Be aware that this operation will overwrite the original text.

```APIDOC
`Translate with 3 Quick Spacebar Taps`: After entering a message in the dialogue interface input box, quickly tapping the spacebar three times can translate the entered content into English.
Note: This operation will overwrite the original text.
```

--------------------------------

### API Error Codes Reference

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/questions-and-feedback/faq

Detailed reference for common HTTP error codes (4xx and 5xx) returned by the API, including possible scenarios and recommended solutions for each code.

```APIDOC
Common Error Codes:
  4xx (Client Error Status Code):
    Description: Generally indicate that the request cannot be completed because of request syntax error, authentication failure, or authentication failure.
  5xx (Server Error Status Code):
    Description: Generally server-side errors, server downtime, request processing timeout, etc.

Detailed Error Codes:
  400:
    Possible Scenarios:
      - Wrong request format.
      - [Common Case 1]: If it is a Gemini model, you may need bind a card.
      - [Common Case 2]: Data volume exceeds the limit. Commonly in Visual Models, the image volume exceeds the upper limit of the upstream single request traffic will return the error code.
      - [Common Case 3]: Add unsupported parameters or fill in the parameters incorrectly. Try to create a new pure assistant to test whether it is normal.
      - [Common Scenario 4]: Context exceeds the limit, clear the context or create a new dialog or reduce the number of context entries.
    Solution:
      - Check the contents of the error returned by the dialog or the console to see what is reported as an error, and follow the prompts.
      - For Gemini model, bind a card.
      - For Visual Models, reduce image volume.
      - Create a new pure assistant to test.
      - Clear context or reduce context entries.
  401:
    Possible Scenarios:
      - Authentication failed: Model not supported or the server account is banned.
    Solution:
      - Contact or check the status of the corresponding service provider's account.
  403:
    Possible Scenarios:
      - Operation not authorized.
    Solution:
      - Act according to the error messages returned by the dialog or the console error message prompts.
  404:
    Possible Scenarios:
      - Resource not found.
    Solution:
      - Checking request paths.
  429:
    Possible Scenarios:
      - Request rate limit reached (TPM or RPM).
    Solution:
      - Try again after a while.
  500:
    Possible Scenarios:
      - Internal server error, unable to complete request.
    Solution:
      - Contact upstream service provider if persistent.
  501:
    Possible Scenarios:
      - The server does not support the requested function and cannot complete the request.
    Solution: ""
  502:
    Possible Scenarios:
      - A server working as a gateway or proxy receives an invalid response from a remote server when it tries to execute a request.
    Solution: ""
  503:
    Possible Scenarios:
      - The server is temporarily unable to process client requests due to overload or system maintenance.
    Solution:
      - The length of the delay can be included in the server's Retry-After header information.
  504:
    Possible Scenarios:
      - Servers acting as gateways or proxies that do not get requests from remote servers in a timely manner.
    Solution: ""
```

--------------------------------

### Configure Assistant Name

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

This setting allows users to customize the name of an assistant for easier identification and management within the assistant interface.

```APIDOC
`Name`: You can customize an assistant name for easy identification.
```

--------------------------------

### Customize Software Appearance with Custom CSS Variables

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/advanced-tutorial/personalization/custom-css

This CSS snippet demonstrates how to customize the software's appearance by overriding default CSS variables within the `:root` selector. It sets various background and chat-related colors. Additionally, it targets the `#content-container` to apply a specific background color, ensuring the changes take effect with `!important`.

```css
:root {
  --color-background: #1a462788;
  --color-background-soft: #1a4627aa;
  --color-background-mute: #1a462766;
  --navbar-background: #1a4627;
  --chat-background: #1a4627;
  --chat-background-user: #28b561;
  --chat-background-assistant: #1a462722;
}

#content-container {
  background-color: #2e5d3a !important;
}
```

--------------------------------

### Configure Display of Estimated Token Count

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

This setting displays an estimated token count for the input text within the input box. This count is for reference only and does not represent the actual tokens consumed by the context.

```APIDOC
`Show Estimated Token Count`: Displays the estimated number of tokens consumed by the input text in the input box (not the actual tokens consumed by the context, for reference only).
```

--------------------------------

### Control Auto Reset Model Behavior

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

Determines whether a new topic resets to the assistant's default model or follows the model used in the previous topic.

```APIDOC
Parameter: Auto Reset Model
  description: Controls model selection for new topics.
  behavior:
    - On: New topic resets to assistant's default model.
    - Off: New topic follows the model used in the previous topic.
  example:
    assistant_default: gpt-3.5-turbo
    topic1_model_switch: gpt-4o
    - If On: New topic 2 defaults to gpt-3.5-turbo.
    - If Off: New topic 2 defaults to gpt-4o.
```

--------------------------------

### Alibaba Cloud Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Alibaba Cloud with their respective maximum input token limits.

```APIDOC
| Name                    | Max Input |
| ----------------------- | --------- |
| text-embedding-v3       | 8192      |
| text-embedding-v2       | 2048      |
| text-embedding-v1       | 2048      |
| text-embedding-async-v2 | 2048      |
| text-embedding-async-v1 | 2048      |
```

--------------------------------

### Set Top P (Nucleus Sampling) for Vocabulary Diversity

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

Affects output by controlling the probability threshold for vocabulary selection, influencing the monotony or diversity of AI-generated content.

```APIDOC
Parameter: Top P (Nucleus Sampling)
  description: Controls vocabulary diversity by setting a probability threshold for selection.
  default_value: 1.0
  range: 0.0 - 1.0
  effects:
    - Smaller value (0.1-0.3):
        output: More conservative and controllable.
        selection: Only considers highest probability vocabulary.
        suitable_for: Code comments, technical documentation.
    - Medium value (0.4-0.6):
        output: Balances vocabulary diversity and accuracy.
        suitable_for: General conversation and writing tasks.
    - Larger value (0.7-1.0):
        output: Richer and more diverse content.
        selection: Considers wider range of vocabulary choices.
        suitable_for: Creative writing, diversified expression.
  notes:
    - Can be used independently or in combination with Temperature.
    - Choose appropriate values based on task type.
    - Experimentation is recommended to find optimal combinations.
    - Parameter ranges are conceptual and may vary by model. Refer to model documentation.
```

--------------------------------

### Adjust Model Temperature for Creativity

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

Controls the randomness and creativity of the text generated by the model, influencing word selection probability.

```APIDOC
Parameter: Temperature
  description: Controls randomness and creativity of generated text.
  default_value: 0.7
  range: 0.0 - 1.0
  effects:
    - Low (0.0-0.3):
        output: More deterministic and focused.
        suitable_for: Code generation, data analysis (requiring accuracy).
        selection: Tends to choose most probable words.
    - Medium (0.4-0.7):
        output: Balances creativity and coherence.
        suitable_for: Daily conversations, general writing (recommended ~0.5 for chatbots).
    - High (0.8-1.0):
        output: More creative and diverse. May reduce text coherence.
        suitable_for: Creative writing, brainstorming.
```

--------------------------------

### Enable Message Length Limit (MaxToken)

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

Sets the maximum number of tokens for a single model response, directly affecting the quality and length of generated output.

```APIDOC
Parameter: Enable Message Length Limit (MaxToken)
  description: Maximum number of tokens for a single response.
  impact: Directly affects quality and length of model's generated responses.
  setting_guidance: Depends on specific needs; refer to suggestions.
  example:
    scenario: Testing model connection in CherryStudio.
    need: Only confirm model returns messages, not specific content.
    setting: Set MaxToken to 1.
  related_concept: Tokens (link: https://docs.cherry-ai.com/question-contact/knowledge#shen-me-shi-tokens)
```

--------------------------------

### Configuring Custom Request Parameters for LLM API Calls

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

This section explains how to add extra request parameters to the API request body. These parameters, such as `presence_penalty`, `top-p`, `maxtokens`, `stream`, and `model`, allow for fine-grained control over model behavior. Custom parameters have higher priority and will override any built-in parameters if duplicated.

```APIDOC
Custom Parameters Structure:
  Description: Add extra request parameters to the request body (body).
  How to fill in: Parameter Name - Parameter Type (Text, Number, etc.) - Value.
  Priority: Custom parameters have higher priority than built-in parameters. If duplicated, custom parameters override built-in ones.

  Example Parameters:
    presence_penalty: (Number) Controls the model's tendency to repeat tokens.
    top-p: (Number) Controls the diversity of the output.
    maxtokens: (Number) Sets the maximum number of tokens to generate.
    stream: (Boolean) Enables or disables stream output.
    model: (Text) Specifies the LLM model to use (e.g., 'gpt-4o').
      Example Usage: Setting 'model' to 'gpt-4o' in custom parameters will force the use of 'gpt-4o' regardless of other dialogue settings.
```

--------------------------------

### Resolving Formula Rendering Issues in CherryStudio

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/questions-and-feedback/faq

This section addresses common problems with formula rendering, such as formulas appearing as raw code instead of formatted output. It emphasizes the correct usage of delimiters for both inline and independent formula blocks and suggests switching to the KateX engine when formulas contain Chinese (CJK) characters to prevent rendering errors.

```LaTeX Syntax
Usage of delimiter:

Inline formula:
- Use a single dollar sign: $formula$
- Or use \\( and \\): \\(formula\\)

Independent formula block:
- Use double dollar symbols: $$formula$$
- Or use \\[formula\\]
- Example: $$\sum_{i=1}^n x_i$$
```

--------------------------------

### OpenAI Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by OpenAI with their respective maximum input token limits.

```APIDOC
| Name                   | Max Input |
| ---------------------- | --------- |
| text-embedding-3-small | 8191      |
| text-embedding-3-large | 8191      |
| text-embedding-ada-002 | 8191      |
```

--------------------------------

### Control Markdown Rendering for Input Messages

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

This setting determines whether messages sent by the user are rendered using Markdown. When turned off, only the model's replies are rendered, while sent messages remain unrendered.

```APIDOC
`Markdown Render Input Messages`: When turned off, only the messages replied by the model are rendered, and the sent messages are not rendered.
```

--------------------------------

### Configure Long Text Pasting Behavior

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/cherrystudio/preview/chat

When a large amount of text is pasted into the input box, this setting automatically formats it as a file-like display. This helps reduce visual clutter and interference when entering subsequent content.

```APIDOC
`Paste Long Text as File`: When a long text is copied from elsewhere and pasted into the input box, it will automatically be displayed as a file style to reduce interference when entering subsequent content.
```

--------------------------------

### Upstage Embedding Models Reference

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Details the embedding models provided by Upstage, including their names and maximum input token limits.

```APIDOC
Upstage Embedding Models:
  embedding-query:
    Max Input: 4000
  embedding-passage:
    Max Input: 4000
```

--------------------------------

### Volcano Engine - Doubao Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Volcano Engine - Doubao with their respective maximum input token limits.

```APIDOC
| Name                    | Max Input |
| ----------------------- | --------- |
| Doubao-embedding        | 4095      |
| Doubao-embedding-vision | 8191      |
| Doubao-embedding-large  | 4095      |
```

--------------------------------

### Baidu Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Baidu with their respective maximum input token limits.

```APIDOC
| Name         | Max Input |
| ------------ | --------- |
| Embedding-V1 | 384       |
| tao-8k       | 8192      |
```

--------------------------------

### Silicon Flow Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Silicon Flow with their respective maximum input token limits.

```APIDOC
| Name                                  | Max Input |
| ------------------------------------- | --------- |
| BAAI/bge-m3                           | 8191      |
| netease-youdao/bce-embedding-base_v1 | 512       |
| BAAI/bge-large-zh-v1.5                | 512       |
| BAAI/bge-large-en-v1.5                | 512       |
| Pro/BAAI/bge-m3                       | 8191      |
```

--------------------------------

### Zhipu AI Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Zhipu AI with their respective maximum input token limits.

```APIDOC
| Name        | Max Input |
| ----------- | --------- |
| embedding-2 | 1024      |
| embedding-3 | 2048      |
```

--------------------------------

### Baichuan Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Baichuan with their respective maximum input token limits.

```APIDOC
| Name                    | Max Input |
| ----------------------- | --------- |
| Baichuan-Text-Embedding | 512       |
```

--------------------------------

### Cohere Embedding Models Reference

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Details the embedding models provided by Cohere, including their names and maximum input token limits for various versions and languages.

```APIDOC
Cohere Embedding Models:
  embed-english-v3.0:
    Max Input: 512
  embed-english-light-v3.0:
    Max Input: 512
  embed-multilingual-v3.0:
    Max Input: 512
  embed-multilingual-light-v3.0:
    Max Input: 512
  embed-english-v2.0:
    Max Input: 512
  embed-english-light-v2.0:
    Max Input: 512
  embed-multilingual-v2.0:
    Max Input: 256
```

--------------------------------

### Together AI Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Together AI with their respective maximum input token limits.

```APIDOC
| Name                      | Max Input |
| ------------------------- | --------- |
| M2-BERT-80M-2K-Retrieval  | 2048      |
| M2-BERT-80M-8K-Retrieval  | 8192      |
| M2-BERT-80M-32K-Retrieval | 32768     |
| UAE-Large-v1              | 512       |
| BGE-Large-EN-v1.5         | 512       |
| BGE-Base-EN-v1.5          | 512       |
```

--------------------------------

### Jina AI Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Jina AI with their respective maximum input token limits.

```APIDOC
| Name                               | Max Input |
| ---------------------------------- | --------- |
| jina-embedding-b-en-v1             | 512       |
| jina-embeddings-v2-base-en         | 8191      |
| jina-embeddings-v2-base-zh         | 8191      |
| jina-embeddings-v2-base-de         | 8191      |
| jina-embeddings-v2-base-code       | 8191      |
| jina-embeddings-v2-base-es         | 8191      |
| jina-colbert-v1-en                 | 8191      |
| jina-reranker-v1-base-en           | 8191      |
| jina-reranker-v1-turbo-en          | 8191      |
| jina-reranker-v1-tiny-en           | 8191      |
| jina-clip-v1                       | 8191      |
| jina-reranker-v2-base-multilingual | 8191      |
| reader-lm-1.5b                     | 256000    |
| reader-lm-0.5b                     | 256000    |
| jina-colbert-v2                    | 8191      |
| jina-embeddings-v3                 | 8191      |
```

--------------------------------

### Gemini Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Gemini with their respective maximum input token limits.

```APIDOC
| Name               | Max Input |
| ------------------ | --------- |
| text-embedding-004 | 2048      |
```

--------------------------------

### Nomic AI Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Nomic AI with their respective maximum input token limits.

```APIDOC
| Name                  | Max Input |
| --------------------- | --------- |
| nomic-embed-text-v1   | 8192      |
| nomic-embed-text-v1.5 | 8192      |
| gte-multilingual-base | 8192      |
```

--------------------------------

### Hunyuan Embedding Models

Source: https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us/knowledge-base/embedding-model-information

Lists embedding models provided by Hunyuan with their respective maximum input token limits.

```APIDOC
| Name              | Max Input |
| ----------------- | --------- |
| hunyuan-embedding | 1024      |
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.