Google Gemini¶

v3¶

Class: GoogleGeminiBlockV3 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.foundation.google_gemini.v3.GoogleGeminiBlockV3

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Ask a question to Google's Gemini model with vision capabilities.

You can specify arbitrary text prompts or predefined ones, the block supports the following types of prompt:

Open Prompt (unconstrained) - Use any prompt to generate a raw response
Text Recognition (OCR) (ocr) - Model recognizes text in the image
Visual Question Answering (visual-question-answering) - Model answers the question you submit in the prompt
Captioning (short) (caption) - Model provides a short description of the image
Captioning (detailed-caption) - Model provides a long description of the image
Single-Label Classification (classification) - Model classifies the image content as one of the provided classes
Multi-Label Classification (multi-label-classification) - Model classifies the image content as one or more of the provided classes
Unprompted Object Detection (object-detection) - Model detects and returns the bounding boxes for prominent objects in the image
Structured Output Generation (structured-answering) - Model returns a JSON response with the specified fields

API Key Options¶

This block supports two API key modes:

Roboflow Managed API Key (Default) - Use rf_key:account to proxy requests through Roboflow's API:
Simplified setup - no Google AI API key required
Secure - your workflow API key is used for authentication
Usage-based billing - charged per token based on the model used
Custom Google AI API Key - Provide your own Google AI API key:
Full control over API usage
You pay Google directly

WARNING!

This block makes use of /v1beta API of Google Gemini model - the implementation may change in the future, without guarantee of backward compatibility.

Type identifier¶

Use the following identifier in step "type" field: roboflow_core/google_gemini@v3to add the block as as step in your workflow.

Properties¶

Name	Type	Description	Refs
`name`	`str`	Enter a unique identifier for this step..	❌
`task_type`	`str`	Task type to be performed by model. Value determines required parameters and output response..	❌
`prompt`	`str`	Text prompt to the Gemini model.	✅
`output_structure`	`Dict[str, str]`	Dictionary with structure of expected JSON response.	❌
`classes`	`List[str]`	List of classes to be used.	✅
`api_key`	`str`	Your Google AI API key or 'rf_key:account' to use Roboflow's managed API key.	✅
`model_version`	`str`	Model to be used.	✅
`thinking_level`	`str`	Controls the depth of internal reasoning for Gemini 3+ models. 'low' minimizes latency and cost (best for simple tasks), 'high' maximizes reasoning depth (default). Only supported by Gemini 3 and newer models..	✅
`temperature`	`float`	Temperature to sample from the model - value in range 0.0-2.0, the higher - the more random / "creative" the generations are..	✅
`max_tokens`	`int`	Maximum number of tokens the model can generate in it's response. If not specified, the model will use its default limit..	❌
`google_code_execution`	`bool`	Enable native code execution for the Gemini model..	✅
`max_concurrent_requests`	`int`	Number of concurrent requests that can be executed by block when batch of input images provided. If not given - block defaults to value configured globally in Workflows Execution Engine. Please restrict if you hit Google Gemini API limits..	❌

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Runtime compatibility¶

requires_internet — air-gapped / offline deployments: This block depends on a service that is not reachable from fully offline / air-gapped deployments.

Available Connections¶

Compatible Blocks

Check what blocks you can connect to Google Gemini in version v3.

Input and Output Bindings¶

The available connections depend on its binding kinds. Check what binding kinds Google Gemini in version v3 has.

Bindings

input
- images (image): The image to infer on..
- prompt (string): Text prompt to the Gemini model.
- classes (list_of_values): List of classes to be used.
- api_key (Union[secret, string, ROBOFLOW_MANAGED_KEY]): Your Google AI API key or 'rf_key:account' to use Roboflow's managed API key.
- model_version (string): Model to be used.
- thinking_level (string): Controls the depth of internal reasoning for Gemini 3+ models. 'low' minimizes latency and cost (best for simple tasks), 'high' maximizes reasoning depth (default). Only supported by Gemini 3 and newer models..
- temperature (float): Temperature to sample from the model - value in range 0.0-2.0, the higher - the more random / "creative" the generations are..
- google_code_execution (boolean): Enable native code execution for the Gemini model..
output
- output (Union[string, language_model_output]): String value if string or LLM / VLM output if language_model_output.
- classes (list_of_values): List of values of any type.

Example JSON definition of step Google Gemini in version v3

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/google_gemini@v3",
    "images": "$inputs.image",
    "task_type": "<block_does_not_provide_example>",
    "prompt": "my prompt",
    "output_structure": {
        "my_key": "description"
    },
    "classes": [
        "class-a",
        "class-b"
    ],
    "api_key": "rf_key:account",
    "model_version": "gemini-3.1-pro-preview",
    "thinking_level": "<block_does_not_provide_example>",
    "temperature": "<block_does_not_provide_example>",
    "max_tokens": "<block_does_not_provide_example>",
    "google_code_execution": "<block_does_not_provide_example>",
    "max_concurrent_requests": "<block_does_not_provide_example>"
}

v2¶

Class: GoogleGeminiBlockV2 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.foundation.google_gemini.v2.GoogleGeminiBlockV2

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Ask a question to Google's Gemini model with vision capabilities.

You can specify arbitrary text prompts or predefined ones, the block supports the following types of prompt:

Open Prompt (unconstrained) - Use any prompt to generate a raw response
Text Recognition (OCR) (ocr) - Model recognizes text in the image
Visual Question Answering (visual-question-answering) - Model answers the question you submit in the prompt
Captioning (short) (caption) - Model provides a short description of the image
Captioning (detailed-caption) - Model provides a long description of the image
Single-Label Classification (classification) - Model classifies the image content as one of the provided classes
Multi-Label Classification (multi-label-classification) - Model classifies the image content as one or more of the provided classes
Unprompted Object Detection (object-detection) - Model detects and returns the bounding boxes for prominent objects in the image
Structured Output Generation (structured-answering) - Model returns a JSON response with the specified fields

You need to provide your Google AI API key to use the Gemini model.

WARNING!

This block makes use of /v1beta API of Google Gemini model - the implementation may change in the future, without guarantee of backward compatibility.

Type identifier¶

Use the following identifier in step "type" field: roboflow_core/google_gemini@v2to add the block as as step in your workflow.

Properties¶

Name	Type	Description	Refs
`name`	`str`	Enter a unique identifier for this step..	❌
`task_type`	`str`	Task type to be performed by model. Value determines required parameters and output response..	❌
`prompt`	`str`	Text prompt to the Gemini model.	✅
`output_structure`	`Dict[str, str]`	Dictionary with structure of expected JSON response.	❌
`classes`	`List[str]`	List of classes to be used.	✅
`api_key`	`str`	Your Google AI API key.	✅
`model_version`	`str`	Model to be used.	✅
`thinking_level`	`str`	Controls the depth of internal reasoning for Gemini 3+ models. 'low' minimizes latency and cost (best for simple tasks), 'high' maximizes reasoning depth (default). Only supported by Gemini 3 and newer models..	✅
`temperature`	`float`	Temperature to sample from the model - value in range 0.0-2.0, the higher - the more random / "creative" the generations are..	✅
`max_tokens`	`int`	Maximum number of tokens the model can generate in it's response. If not specified, the model will use its default limit..	❌
`max_concurrent_requests`	`int`	Number of concurrent requests that can be executed by block when batch of input images provided. If not given - block defaults to value configured globally in Workflows Execution Engine. Please restrict if you hit Google Gemini API limits..	❌

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Runtime compatibility¶

requires_internet — air-gapped / offline deployments: This block depends on a service that is not reachable from fully offline / air-gapped deployments.

Available Connections¶

Compatible Blocks

Check what blocks you can connect to Google Gemini in version v2.

Input and Output Bindings¶

The available connections depend on its binding kinds. Check what binding kinds Google Gemini in version v2 has.

Bindings

input
- images (image): The image to infer on..
- prompt (string): Text prompt to the Gemini model.
- classes (list_of_values): List of classes to be used.
- api_key (Union[secret, string]): Your Google AI API key.
- model_version (string): Model to be used.
- thinking_level (string): Controls the depth of internal reasoning for Gemini 3+ models. 'low' minimizes latency and cost (best for simple tasks), 'high' maximizes reasoning depth (default). Only supported by Gemini 3 and newer models..
- temperature (float): Temperature to sample from the model - value in range 0.0-2.0, the higher - the more random / "creative" the generations are..
output
- output (Union[string, language_model_output]): String value if string or LLM / VLM output if language_model_output.
- classes (list_of_values): List of values of any type.

Example JSON definition of step Google Gemini in version v2

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/google_gemini@v2",
    "images": "$inputs.image",
    "task_type": "<block_does_not_provide_example>",
    "prompt": "my prompt",
    "output_structure": {
        "my_key": "description"
    },
    "classes": [
        "class-a",
        "class-b"
    ],
    "api_key": "xxx-xxx",
    "model_version": "gemini-3.1-pro-preview",
    "thinking_level": "<block_does_not_provide_example>",
    "temperature": "<block_does_not_provide_example>",
    "max_tokens": "<block_does_not_provide_example>",
    "max_concurrent_requests": "<block_does_not_provide_example>"
}

v1¶

Class: GoogleGeminiBlockV1 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.foundation.google_gemini.v1.GoogleGeminiBlockV1

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Ask a question to Google's Gemini model with vision capabilities.

You can specify arbitrary text prompts or predefined ones, the block supports the following types of prompt:

Open Prompt (unconstrained) - Use any prompt to generate a raw response
Text Recognition (OCR) (ocr) - Model recognizes text in the image
Visual Question Answering (visual-question-answering) - Model answers the question you submit in the prompt
Captioning (short) (caption) - Model provides a short description of the image
Captioning (detailed-caption) - Model provides a long description of the image
Single-Label Classification (classification) - Model classifies the image content as one of the provided classes
Multi-Label Classification (multi-label-classification) - Model classifies the image content as one or more of the provided classes
Unprompted Object Detection (object-detection) - Model detects and returns the bounding boxes for prominent objects in the image
Structured Output Generation (structured-answering) - Model returns a JSON response with the specified fields

You need to provide your Google AI API key to use the Gemini model.

WARNING!

This block makes use of /v1beta API of Google Gemini model - the implementation may change in the future, without guarantee of backward compatibility.

Type identifier¶

Use the following identifier in step "type" field: roboflow_core/google_gemini@v1to add the block as as step in your workflow.

Properties¶

Name	Type	Description	Refs
`name`	`str`	Enter a unique identifier for this step..	❌
`task_type`	`str`	Task type to be performed by model. Value determines required parameters and output response..	❌
`prompt`	`str`	Text prompt to the Gemini model.	✅
`output_structure`	`Dict[str, str]`	Dictionary with structure of expected JSON response.	❌
`classes`	`List[str]`	List of classes to be used.	✅
`api_key`	`str`	Your Google AI API key.	✅
`model_version`	`str`	Model to be used.	✅
`max_tokens`	`int`	Maximum number of tokens the model can generate in it's response..	❌
`temperature`	`float`	Temperature to sample from the model - value in range 0.0-2.0, the higher - the more random / "creative" the generations are..	✅
`max_concurrent_requests`	`int`	Number of concurrent requests that can be executed by block when batch of input images provided. If not given - block defaults to value configured globally in Workflows Execution Engine. Please restrict if you hit Google Gemini API limits..	❌

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Runtime compatibility¶

requires_internet — air-gapped / offline deployments: This block depends on a service that is not reachable from fully offline / air-gapped deployments.

Available Connections¶

Compatible Blocks

Check what blocks you can connect to Google Gemini in version v1.

Input and Output Bindings¶

The available connections depend on its binding kinds. Check what binding kinds Google Gemini in version v1 has.

Bindings

input
- images (image): The image to infer on..
- prompt (string): Text prompt to the Gemini model.
- classes (list_of_values): List of classes to be used.
- api_key (Union[secret, string]): Your Google AI API key.
- model_version (string): Model to be used.
- temperature (float): Temperature to sample from the model - value in range 0.0-2.0, the higher - the more random / "creative" the generations are..
output
- output (Union[string, language_model_output]): String value if string or LLM / VLM output if language_model_output.
- classes (list_of_values): List of values of any type.

Example JSON definition of step Google Gemini in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/google_gemini@v1",
    "images": "$inputs.image",
    "task_type": "<block_does_not_provide_example>",
    "prompt": "my prompt",
    "output_structure": {
        "my_key": "description"
    },
    "classes": [
        "class-a",
        "class-b"
    ],
    "api_key": "xxx-xxx",
    "model_version": "gemini-2.5-pro",
    "max_tokens": "<block_does_not_provide_example>",
    "temperature": "<block_does_not_provide_example>",
    "max_concurrent_requests": "<block_does_not_provide_example>"
}