Configure Google Gemini

Google Gemini is a multimodal AI model series developed by Google, supporting various input forms such as text, images, and audio. It provides multi-tier model choices from lightweight to flagship, with powerful reasoning, analysis, and creative capabilities.

1. Obtain Google Gemini API Key

1.1 Visit AI Studio

Visit AI Studio and log in: https://aistudio.google.com/

1.2 Enter API Keys Page

After logging in, click Get API key in the left menu.

1.3 Create New API Key

Click the Create API key button.

1.4 Select Google Cloud Project

In the pop-up dialog:

1. Select an existing project or create a new project
2. Click Create API key in existing project or Create API key in new project

1.5 Copy API Key

After successful creation, the system will display the API Key.

Important: Please copy and save it immediately. The API Key starts with AIzaSy and is 39 characters long.

Click the copy button, and the API Key is copied to the clipboard.

2. Configure Gemini Model in CueMate

2.1 Enter Model Settings Page

After logging into CueMate, click Model Settings in the dropdown menu in the upper right corner.

2.2 Add New Model

Click the Add Model button in the upper right corner.

2.3 Select Gemini Provider

In the pop-up dialog:

1. Provider Type: Select Gemini
2. After clicking, automatically proceed to the next step

2.4 Fill in Configuration Information

Fill in the following information on the configuration page:

Basic Configuration

1. Model Name: Give this model configuration a name (e.g., Gemini-2.0-Flash)
2. API Key: Paste the Google Gemini API Key you just copied
3. API URL: Keep the default https://generativelanguage.googleapis.com/v1beta
4. Model Version: Select or enter the model you want to use (recommended to use full version number)
   • gemini-2.5-pro-exp-03-25: 2.5 Pro experimental version, best performance
   • gemini-2.0-flash-001: 2.0 Flash stable version, recommended
   • gemini-2.0-flash-exp: 2.0 Flash experimental version
   • gemini-1.5-pro-latest: 1.5 Pro latest version
   • gemini-1.5-pro-002: 1.5 Pro stable version
   • gemini-1.5-flash-latest: 1.5 Flash latest version
   • gemini-1.5-flash-002: 1.5 Flash stable version
   • gemini-1.5-flash-8b: 8B parameter lightweight version
   • gemini-1.0-pro: 1.0 Pro standard version (legacy)
   • gemini-1.0-pro-vision: 1.0 with vision understanding (legacy)

Advanced Configuration (Optional)

Expand the Advanced Configuration panel to adjust the following parameters:

Parameters Adjustable in CueMate Interface:

1. Temperature: Controls output randomness

   • Range: 0-2
   • Recommended Value: 0.7
   • Function: Higher values produce more random and creative output, lower values produce more stable and conservative output
   • Usage Suggestions:
     • Creative writing/brainstorming: 1.0-1.5
     • Regular conversation/Q&A: 0.7-0.9
     • Code generation/precise tasks: 0.3-0.5
   • Note: Gemini's temperature range is 0-2, different from Claude's 0-1

2. Max Output Tokens (maxOutputTokens): Limits single output length

   • Range: 256 - 65536 (depending on model)
   • Recommended Value: 8192
   • Function: Controls the maximum word count of model's single response
   • Model Limits:
     • Gemini 2.5 Pro: Maximum 65K tokens
     • Gemini 2.0 Series: Maximum 8K tokens
     • Gemini 1.5 Series: Maximum 8K tokens
     • Gemini 1.0 Pro: Maximum 2K tokens
     • Gemini 1.0 Pro Vision: Maximum 4K tokens
   • Usage Suggestions:
     • Short Q&A: 1024-2048
     • Regular conversation: 4096-8192
     • Long text generation: 16384-32768
     • Ultra-long output: 65536 (2.5 Pro only)

Other Advanced Parameters Supported by Google Gemini API:

Although the CueMate interface only provides temperature and maxOutputTokens adjustments, if you call Gemini directly through the API, you can also use the following advanced parameters:

3. topP (nucleus sampling)

   • Range: 0-1
   • Default Value: 0.95
   • Function: Samples from the minimum candidate set where cumulative probability reaches p
   • Relationship with temperature: Can be used simultaneously
   • Usage Suggestions:
     • Maintain diversity: 0.9-1.0
     • More conservative output: 0.7-0.8

4. topK

   • Range: 1-40
   • Default Value: 40 (or null, meaning no limit)
   • Function: Samples from the k candidate words with the highest probability
   • Usage Suggestions:
     • More diverse: 32-40
     • More conservative: 10-20
   • Note: Gemini's topK maximum value is 40

5. stopSequences (stop sequences)

   • Type: String array
   • Default Value: null
   • Maximum Count: 5 strings
   • Function: Stops when generated content contains specified strings
   • Example: ["END", "###", "\n\n"]
   • Use Cases:
     • Structured output: Use delimiters to control format
     • Dialogue system: Prevent model from continuing to generate
     • Code block control: Use code block end markers

6. candidateCount

   • Type: Integer
   • Range: 1-8
   • Default Value: 1
   • Function: Generate multiple candidate responses for selection
   • Use Cases:
     • A/B testing different outputs
     • Selecting best response
   • Note: Will increase API call cost

Gemini Safety Settings:

7. safetySettings
   • Type: Object array
   • Function: Configure content safety filtering level
   • Safety Categories:
     • HARM_CATEGORY_HARASSMENT: Harassment content
     • HARM_CATEGORY_HATE_SPEECH: Hate speech
     • HARM_CATEGORY_SEXUALLY_EXPLICIT: Sexually explicit content
     • HARM_CATEGORY_DANGEROUS_CONTENT: Dangerous content
   • Filtering Levels:
     • BLOCK_NONE: No filtering
     • BLOCK_LOW_AND_ABOVE: Filter low risk and above
     • BLOCK_MEDIUM_AND_ABOVE: Filter medium and above (default)
     • BLOCK_ONLY_HIGH: Only filter high risk
   • Example:

     {
       "safetySettings": [
         {
           "category": "HARM_CATEGORY_HARASSMENT",
           "threshold": "BLOCK_MEDIUM_AND_ABOVE"
         }
       ]
     }

   • Usage Suggestions:
     • Public-facing applications: Keep default settings
     • Internal tools/research: Can relax appropriately (BLOCK_ONLY_HIGH)
     • Strict scenarios: Use BLOCK_LOW_AND_ABOVE

Scenario	temperature	maxOutputTokens	topP	topK	stopSequences
Creative Writing	1.0-1.2	4096-8192	0.95	40	null
Code Generation	0.2-0.5	2048-4096	0.9	30	null
Q&A System	0.7	1024-2048	0.9	40	null
Summary	0.3-0.5	512-1024	0.9	20	null
Structured Output	0.5	2048	0.9	30	["END", "###"]

2.5 Test Connection

After filling in the configuration, click the Test Connection button to verify if the configuration is correct.

If the configuration is correct, a successful test message will be displayed, along with a sample response from the model.

If the configuration is incorrect, test error logs will be displayed, and you can view specific error information through log management.

2.6 Save Configuration

After a successful test, click the Save button to complete the model configuration.

3. Use Model

Through the dropdown menu in the upper right corner, enter the system settings interface, and select the model configuration you want to use in the LLM provider section.

After configuration, you can select to use this model in interview training, question generation, and other functions. Of course, you can also select the model configuration for a specific interview in the interview options.

4. Supported Model List

4.1 Gemini 2.5 Series (2025 Latest)

No.	Model Name	Model ID	Max Output	Use Case
1	Gemini 2.5 Pro (Experimental)	gemini-2.5-pro-exp-03-25	65K tokens	Best performance, long text generation

4.2 Gemini 2.0 Series

No.	Model Name	Model ID	Max Output	Use Case
1	Gemini 2.0 Flash	gemini-2.0-flash-001	8K tokens	Recommended, stable version
2	Gemini 2.0 Flash (Experimental)	gemini-2.0-flash-exp	8K tokens	Experimental feature testing

4.3 Gemini 1.5 Series

No.	Model Name	Model ID	Max Output	Use Case
1	Gemini 1.5 Pro (Latest)	gemini-1.5-pro-latest	8K tokens	Automatically uses latest version
2	Gemini 1.5 Pro 002	gemini-1.5-pro-002	8K tokens	Complex reasoning, technical interviews
3	Gemini 1.5 Flash (Latest)	gemini-1.5-flash-latest	8K tokens	Automatically uses latest version
4	Gemini 1.5 Flash 002	gemini-1.5-flash-002	8K tokens	Fast response, regular conversation
5	Gemini 1.5 Flash 8B	gemini-1.5-flash-8b	8K tokens	Lightweight tasks, low cost

4.4 Gemini 1.0 Series (Legacy)

No.	Model Name	Model ID	Max Output	Use Case
1	Gemini 1.0 Pro	gemini-1.0-pro	2K tokens	Basic conversation
2	Gemini 1.0 Pro Vision	gemini-1.0-pro-vision	4K tokens	Image understanding, multimodal

4.5 Recommended Selection

• Production Environment: gemini-2.0-flash-001 (stable, cost-effective)
• Long Text Generation: gemini-2.5-pro-exp-03-25 (maximum 65K output)
• Fast Response: gemini-1.5-flash-002 (low latency)
• Lightweight Applications: gemini-1.5-flash-8b (lowest cost)

5. Common Issues

5.1 Model Name Does Not Exist (404 Error)

Symptom: models/gemini-1.5-pro is not found for API version v1beta when testing connection

Solution:

1. Check if the model name uses the full version number
2. Wrong Example: gemini-1.5-pro (no version number, will return 404)
3. Correct Examples:
   • gemini-2.0-flash-001 (recommended)
   • gemini-1.5-pro-002
   • gemini-1.5-flash-002
   • gemini-1.5-pro-latest (using latest alias)
4. Do not use short aliases like gemini-1.5-pro or gemini-1.5-flash, must use full version number

5.2 Invalid API Key

Symptom: API Key error when testing connection

Solution:

1. Check if the API Key starts with AIzaSy
2. Confirm the API Key is 39 characters long
3. Check if the Google Cloud project has enabled Generative Language API
4. Confirm the API Key has not been restricted or disabled

5.3 Request Timeout

Symptom: No response for a long time during testing or use

Solution:

1. Check if the network connection is normal
2. If in China, may need to use a proxy
3. Check firewall settings

5.4 Insufficient Quota

Symptom: Quota exhausted or request frequency too high message

Solution:

1. Log in to Google Cloud Console to check quota usage
2. Apply to increase quota limit
3. Optimize usage frequency

5.5 API Not Enabled

Symptom: API not enabled or project has no permission message

Solution:

1. Visit Google Cloud Console
2. Enable Generative Language API
3. Confirm project configuration is correct

5.6 Cannot Access AI Studio (Permission Denied)

Symptom: "permission denied" error immediately after clicking "Get API Key" and redirects to documentation page

Solution:

1. IP Region Restrictions: AI Studio has access restrictions for some regions, need to use a proxy from supported regions like the US
2. Bypass AI Studio to Create API Key:
   • Visit Google Cloud Console - Credentials Page
   • Click "Create Credentials" → "API Key"
   • Copy the generated API Key (starts with AIzaSy)
   • Select "Generative Language API" in "API Restrictions"
3. Note: API Keys created through Cloud Console are identical to those created through AI Studio and can be used normally

Related Links

• Google AI Studio
• Google Cloud Console
• API Documentation
• Pricing Information