Generate an Answer with Gemini 2.5 Flash

Send a current question, a behavior instruction and one or two prior messages to Gemini 2.5 Flash through DataForSEO, with web search enabled. Receive a generated answer and available reference information.

Your AI

Use with your AI

For tasks with several steps, continue in the AI you already use.

CodexClaude CodeCursorWindsurfCline

View instructions

Use Toolmu’s gemini-llm-responses-live for the user’s current question. Reuse the existing Toolmu connection and follow the active authorization and budget. Ask only for the missing question; ask about audience or tone only when it would change the answer. For a new question, no previous chat is needed. If the user supplies real prior conversation, include up to two relevant actual turns with their real user/ai roles; never invent history. Because this Tool requires a nonempty message_chain, when there is no real prior turn, use the current user question as its single user message and also send that question as user_prompt; do not describe it as past conversation. Use a clear, beginner-friendly system_message by default, or follow the user’s requested style. The model, web search and generation settings are fixed; do not ask the user to choose them. Present the actual answer, reported usage and available references. Sources may be absent and search does not guarantee factual accuracy or freshness; identify uncertainty and do not claim citations that were not returned.

Task:
Explain [how solar panels turn sunlight into electricity] for [someone with no science background]. Use everyday language, and list sources if any are returned.

Paste into your connected AI's conversation.

Connection guide

Up to 0.11 credits / call

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

DataForSEO

Generate an Answer with Gemini 2.5 Flash

You receive

Get a Gemini 2.5 Flash response containing generated answer text and reported model and token usage. When available, references include source titles, direct links and passages associated with the answer. The result can also report whether search was used and related queries derived from your prompt. Your Agent can present the answer and available references together so you can review the explanation and check its sources.

Start with your task

Explain [how solar panels turn sunlight into electricity] for [someone with no science background]. Use everyday language, and list sources if any are returned.

What to know

A generated answer can be incomplete or incorrect. Enabled search does not guarantee citations or current facts, and references may be empty. The required 200-token setting is a request, not an exact length or a strict cutoff when search is used. Inspect the actual text before relying on it.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Generation request`request`	`object`	Required	Required object containing all eight properties. Supply the current task, behavior instruction, prior messages and every fixed setting together. Required fields `system_message, message_chain, max_output_tokens, temperature, top_p, model_name, web_search, user_prompt`
Token sampling setting`request.top_p`	`number`	Required	Required number fixed at 0.5. Controls the diversity of token selection; it does not guarantee a deterministic answer. Fixed value `0.5`
Gemini model`request.model_name`	`string`	Required	Required string fixed at gemini-2.5-flash. This Tool does not accept a different model name. Fixed value `gemini-2.5-flash`
Enable web search`request.web_search`	`boolean`	Required	Required boolean fixed at true. Permits web retrieval and citations, but actual search use and references depend on the response. Fixed value `true`
Response randomness setting`request.temperature`	`number`	Required	Required number fixed at 0.3. A lower value favors a more focused response; it is not a guarantee of identical text. Fixed value `0.3`
Current question or task`request.user_prompt`	`string`	Required	Required 1–256-character string containing the task for this response. Supply it alongside the behavior instruction and required prior messages. Minimum length `1` Maximum length `256`
Prior conversation`request.message_chain`	`array`	Required	Required array of 1–2 previous message entries. Use actual supplied conversation context; it does not retrieve an account’s conversation history automatically. Minimum items `1` Maximum items `2`
Prior message entry`request.message_chain[]`	`object`	—	Each entry must contain role and message, both 1–256-character strings. Additional properties are not accepted. Required fields `role, message`
Prior speaker`request.message_chain[].role`	`string`	—	Required 1–256-character string. Use the supplier-compatible role user for a user message or ai for a model reply. Minimum length `1` Maximum length `256`
Prior message text`request.message_chain[].message`	`string`	—	Required 1–256-character string containing the content of this earlier turn. Keep it separate from the current task. Minimum length `1` Maximum length `256`
Behavior instruction`request.system_message`	`string`	Required	Required 1–256-character string describing the model’s role, tone or behavior, such as “Explain clearly for a beginner.” Minimum length `1` Maximum length `256`
Requested output tokens`request.max_output_tokens`	`integer`	Required	Required integer fixed at 200. Tokens are not words or characters. With web search enabled, the returned output token count can exceed this requested limit. Fixed value `200`

View raw JSON

{
  "type": "object",
  "required": [
    "request"
  ],
  "properties": {
    "request": {
      "type": "object",
      "required": [
        "system_message",
        "message_chain",
        "max_output_tokens",
        "temperature",
        "top_p",
        "model_name",
        "web_search",
        "user_prompt"
      ],
      "properties": {
        "top_p": {
          "type": "number",
          "const": 0.5
        },
        "model_name": {
          "type": "string",
          "const": "gemini-2.5-flash"
        },
        "web_search": {
          "type": "boolean",
          "const": true
        },
        "temperature": {
          "type": "number",
          "const": 0.3
        },
        "user_prompt": {
          "type": "string",
          "maxLength": 256,
          "minLength": 1
        },
        "message_chain": {
          "type": "array",
          "items": {
            "type": "object",
            "required": [
              "role",
              "message"
            ],
            "properties": {
              "role": {
                "type": "string",
                "maxLength": 256,
                "minLength": 1
              },
              "message": {
                "type": "string",
                "maxLength": 256,
                "minLength": 1
              }
            },
            "additionalProperties": false
          },
          "maxItems": 2,
          "minItems": 1
        },
        "system_message": {
          "type": "string",
          "maxLength": 256,
          "minLength": 1
        },
        "max_output_tokens": {
          "type": "integer",
          "const": 200
        }
      },
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}

Result schema

Output fields have not been documented yet.

Additional notes

See DataForSEO’s Gemini LLM Responses documentation for the distinction between generated text, source annotations and reported token usage.

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

This Tool uses gemini-2.5-flash with web search enabled, temperature 0.3, top_p 0.5 and a requested output limit of 200 tokens. All settings must be supplied with those fixed values. Lower sampling settings do not guarantee identical answers.

Generate an Answer with Gemini 2.5 Flash

Generate an Answer with Gemini 2.5 Flash

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Connect your AI assistant

Ask in your own words

Read the answer and sources

Questions about this tool.

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Connect your AI assistant

Ask in your own words

Read the answer and sources

Questions about this tool.

Can I choose the model or generation settings?

How do I supply the current task and earlier messages?

Will it always search the web and provide fresh citations?

Does a structured response mean the answer is valid JSON?

Is the answer limited to exactly 200 tokens?