---
title: Gemini 3.1 Flash Lite | Runware Docs
url: https://runware.ai/docs/models/google-gemini-3-1-flash-lite
description: Advanced multimodal text and reasoning model
---
# Gemini 3.1 Flash Lite

![Gemini 3.1 Flash Lite](https://assets.runware.ai/covers/google-gemini-3-1-flash-lite.jpg)

Gemini 3.1 Flash Lite is Google’s flagship multimodal language model that processes text alongside images, audio, video, code, and documents. It offers high-performance reasoning, complex instruction following, and deep contextual understanding for a wide range of tasks across language, analysis, and problem solving

- **Status**: live
- **ID**: `google-gemini-3-1-flash-lite`
- Try in Playground
- **Context window**: 1M context
- **Max output**: 64K output
- **Creator**: Google
- **Release Date**: March 3, 2026
- **Capabilities**: Text to Text, Image to Text, Video to Text, Audio to Text, Checkpoint

## Pricing

~$0.00177 typical run

**Rates** (per 1M tokens)

- text, image, video: `$0.25`
- audio: `$0.50`
- Output: `$1.50`

**5 example generations**

- [tool-calling](https://runware.ai/docs/models/google-gemini-3-1-flash-lite/examples#example-museum-exhibition-readiness-check): `$0.000053` (~1s)
- [text-generation](https://runware.ai/docs/models/google-gemini-3-1-flash-lite/examples#example-cold-chain-incident-briefing): `$0.000652` (~2s)
- [audio-driven](https://runware.ai/docs/models/google-gemini-3-1-flash-lite/examples#example-wholesale-buyer-voicemail-action-brief): `$0.00177` (~7s)
- [structured-output](https://runware.ai/docs/models/google-gemini-3-1-flash-lite/examples#example-retail-lease-abstract-json-record): `$0.00181` (~3s)
- [reference-to-text](https://runware.ai/docs/models/google-gemini-3-1-flash-lite/examples#example-hotel-suite-conformance-report): `$0.00183` (~11s)

billed per token, so a request costs what the model produced · times vary, measured end to end · [How pricing works](https://runware.ai/docs/models-api/pricing)

## Compatibility & Validation

When `jsonSchema` is provided, `outputFormat` must be `JSON`.

## Request Parameters

**API Options**

Platform-level options for task execution and delivery.

### [taskType](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-tasktype)

- **Type**: `string`
- **Required**: true
- **Value**: `textInference`

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-taskuuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.

### [outputFormat](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-outputformat)

- **Type**: `string`
- **Required**: conditional — Required when jsonSchema is provided
- **Default**: `TEXT`

Specifies the file format of the generated output. The available values depend on the task type and the specific model's capabilities.

**Allowed values**: `TEXT` `JSON`

### [webhookURL](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-webhookurl)

- **Type**: `string`
- **Format**: `uri`

Webhook URL that receives JSON responses via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.

**Learn more** (1 resource):

- [Webhooks](https://runware.ai/docs/models-api/webhooks) (platform)

### [deliveryMethod](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-deliverymethod)

- **Type**: `string`
- **Default**: `sync`

Determines how the API delivers task results.

**Allowed values**:

- `sync` Returns results directly in the API response. Over HTTP, a task still running after 120 seconds continues as an async task and comes back with status 'processing'. Poll for its results using getResponse.
- `async` Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse.
- `stream` Streams results token-by-token as they are generated.

**Learn more** (1 resource):

- [Task Polling](https://runware.ai/docs/models-api/task-polling) (platform)

### [includeCost](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [includeUsage](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-includeusage)

- **Type**: `boolean`

Include token usage statistics in the response.

### [numberResults](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-numberresults)

- **Type**: `integer`
- **Min**: `1`
- **Max**: `4`
- **Default**: `1`

Number of results to generate. Each result uses a different seed, producing variations of the same parameters.

**Inputs**

Input resources for the task (images, audio, etc). These must be nested inside the \`inputs\` object.

### [audios](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-inputs-audios)

- **Path**: `inputs.audios`
- **Type**: `array of strings`
- **Min items**: `1`

Array of audio inputs (UUID, URL, or Base64).

### [documents](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-inputs-documents)

- **Path**: `inputs.documents`
- **Type**: `array of strings`
- **Min items**: `1`

Array of document inputs (UUID, URL, or Base64).

### [images](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-inputs-images)

- **Path**: `inputs.images`
- **Type**: `array of strings`
- **Min items**: `1`

Array of image inputs (UUID, URL, Data URI, or Base64).

### [videos](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-inputs-videos)

- **Path**: `inputs.videos`
- **Type**: `array of strings`
- **Min items**: `1`

Array of video inputs (UUID, URL, or Base64).

**Core Parameters**

Primary parameters that define the task output.

### [model](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `google-gemini-3-1-flash-lite`

Send `auto` instead of an AIR identifier to have the model router choose the model for this task. The rest of the request is unchanged, every value you send is kept, and only the parameters you leave out are filled in. See [Model Router](https://runware.ai/docs/models-api/model-router).

**Learn more** (3 resources):

- [Text To Image: Model Selection](https://runware.ai/docs/learn/text-to-image#model-selection) (learn)
- [Image Inpainting: Model Specialized Inpainting Models](https://runware.ai/docs/learn/image-inpainting#model-specialized-inpainting-models) (learn)
- [Image Outpainting: Other Critical Parameters](https://runware.ai/docs/learn/image-outpainting#other-critical-parameters) (learn)

### [seed](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-seed)

- **Type**: `integer`
- **Min**: `0`
- **Max**: `4294967295`

Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.

### [jsonSchema](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-jsonschema)

- **Type**: `object | string`

JSON Schema for structured output. Only honoured when `outputFormat` is JSON. Accepts the OpenAI envelope ({name, schema, strict}) or a bare JSON Schema; bare schemas are auto-wrapped with `name='response'` and `strict=true`.

### [messages](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-messages)

- **Path**: `messages.role`
- **Type**: `array of objects (2 properties)`
- **Required**: true
- **Min items**: `1`

Array of chat messages forming the conversation context. The final message must use the user role.

#### [role](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-messages-role)

- **Path**: `messages.role`
- **Type**: `string`
- **Required**: true

Role of the message author.

**Allowed values**: `user` `assistant`

#### [content](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-messages-content)

- **Path**: `messages.content`
- **Type**: `string`
- **Required**: true
- **Min**: `1`

Text content of the message.

**Settings**

Technical parameters to fine-tune the inference process. These must be nested inside the \`settings\` object.

### [systemPrompt](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-settings-systemprompt)

- **Path**: `settings.systemPrompt`
- **Type**: `string`
- **Min**: `1`
- **Max**: `200000`

System-level instruction that guides the model's behavior and output style across the entire generation.

### [temperature](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-settings-temperature)

- **Path**: `settings.temperature`
- **Type**: `float`
- **Min**: `0`
- **Max**: `2`
- **Step**: `0.01`
- **Default**: `1`

Controls randomness in generation. Lower values produce more deterministic outputs, higher values increase variation and creativity.

### [topP](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-settings-topp)

- **Path**: `settings.topP`
- **Type**: `float`
- **Min**: `0`
- **Max**: `1`
- **Step**: `0.01`
- **Default**: `0.95`

Nucleus sampling parameter that controls diversity by limiting the probability mass. Lower values make outputs more focused, higher values increase diversity.

### [maxTokens](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-settings-maxtokens)

- **Path**: `settings.maxTokens`
- **Type**: `integer`
- **Min**: `1`
- **Max**: `65536`
- **Default**: `4096`

Maximum number of tokens to generate in the response.

### [stopSequences](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-settings-stopsequences)

- **Path**: `settings.stopSequences`
- **Type**: `array of strings`
- **Min**: `1`
- **Max**: `50`
- **Max items**: `5`

Array of sequences that will cause the model to stop generating further tokens when encountered.

### [thinkingLevel](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-settings-thinkinglevel)

- **Path**: `settings.thinkingLevel`
- **Type**: `string`
- **Default**: `high`

Controls the depth of internal reasoning the model performs before generating a response.

**Allowed values**: `minimal` `low` `medium` `high`

### [toolChoice](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-toolchoice)

- **Path**: `toolChoice.type`
- **Type**: `object (2 properties)`

Controls how the model selects which tool to call. This only takes effect when `tools` are defined.

**Examples**:

**Let the model decide (default):**

```json
"toolChoice": {
  "type": "auto"
}
```

**Force a specific tool call:**

```json
"toolChoice": {
  "type": "tool",
  "name": "get_weather"
}
```

**Require any tool call:**

```json
"toolChoice": {
  "type": "any"
}
```

#### [type](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-toolchoice-type)

- **Path**: `toolChoice.type`
- **Type**: `string`
- **Required**: true

Strategy the model uses to decide when and which tools to call.

**Allowed values**:

- `auto` Tool calls decided by the model from the conversation context. This is the recommended default.
- `any` At least one tool call, with the model choosing which tool. Useful when you always need structured output.
- `tool` Call to the specific tool identified by `name`. Use this to force a particular function call.
- `none` No tool calls, even if tools are defined. Useful for forcing a text-only response.

#### [name](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-toolchoice-name)

- **Path**: `toolChoice.name`
- **Type**: `string`

Name of the specific tool the model must call. Required when type is `tool`.

### [tools](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-tools)

- **Path**: `tools.type`
- **Type**: `array of objects (4 properties)`
- **Min items**: `1`

An array of tool definitions that the model may call during generation. The model can invoke one or more tools based on the conversation context, outputting structured calls with arguments instead of (or alongside) free-text.

For `function` tools, each definition requires:

- **`type`**: `"function"`
- **`name`**: Unique identifier (alphanumeric, hyphens, underscores; max 64 chars).
- **`description`**: What the function does. The model uses this to decide when to call it.
- **`schema`**: JSON Schema object describing the expected input arguments.

> [!NOTE]
> The `search` tool is executed server-side by the provider. You don't need to handle the tool result yourself.

> [!NOTE]
> The `codeInterpreter` tool is executed server-side by the provider. You don't need to handle the tool result yourself.

**Examples**:

**Function tool, weather lookup:**

```json
"tools": [
  {
    "type": "function",
    "name": "get_weather",
    "description": "Get current weather for a city",
    "schema": {
      "type": "object",
      "properties": {
        "city": { "type": "string", "description": "City name" }
      },
      "required": ["city"]
    }
  }
],
"toolChoice": { "type": "auto" }
```

**Built-in web search:**

```json
"tools": [
  { "type": "search" }
]
```

**Built-in code interpreter:**

```json
"tools": [
  { "type": "codeInterpreter" }
]
```

**Multiple function tools:**

```json
"tools": [
  {
    "type": "function",
    "name": "search_products",
    "description": "Search the product catalog by query and filters.",
    "schema": {
      "type": "object",
      "properties": {
        "query": { "type": "string" },
        "category": { "type": "string" }
      },
      "required": ["query"]
    }
  },
  {
    "type": "function",
    "name": "add_to_cart",
    "description": "Add a product to the user's shopping cart.",
    "schema": {
      "type": "object",
      "properties": {
        "productId": { "type": "string" },
        "quantity": { "type": "integer", "minimum": 1 }
      },
      "required": ["productId"]
    }
  }
]
```

#### [type](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-tools-type)

- **Path**: `tools.type`
- **Type**: `string`
- **Required**: true

Kind of tool to make available to the model. User-defined functions require `name` and `schema`, while built-in tools (`search`, `codeInterpreter`) are executed server-side by the provider.

**Allowed values**:

- `function` User-defined function tool. The model outputs the tool name and arguments. You execute the function locally and send results back.
- `search` Built-in web search. The provider executes search server-side and enriches the response automatically.
- `codeInterpreter` Built-in code execution sandbox (Python). The provider runs code server-side and returns results automatically.

#### [name](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-tools-name)

- **Path**: `tools.name`
- **Type**: `string`
- **Max**: `64`

Unique function name. Required for function tools.

#### [description](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-tools-description)

- **Path**: `tools.description`
- **Type**: `string`

Explanation of what the function does, used by the model to decide when to call it.

#### [schema](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#request-tools-schema)

- **Path**: `tools.schema`
- **Type**: `object`

JSON Schema object describing the function's input parameters.

## Response Parameters

### [taskType](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-tasktype)

- **Type**: `string`
- **Required**: true
- **Value**: `textInference`

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-taskuuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID v4 identifier echoed from the original request, used to match async responses to their tasks.

### [text](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-text)

- **Type**: `string`
- **Required**: true

Generated text content.

### [cost](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-cost)

- **Type**: `float`

Task cost in USD. Present when `includeCost` is set to `true` in the request.

### [finishReason](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-finishreason)

- **Type**: `string`
- **Required**: true

The reason why the model stopped generating tokens.

**Possible values**: `stop` `length` `content_filter` `unknown`

### [usage](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-usage)

- **Path**: `usage.promptTokens`
- **Type**: `object (4 properties)`
- **Required**: true

Token usage statistics for the request.

#### [promptTokens](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-usage-prompttokens)

- **Path**: `usage.promptTokens`
- **Type**: `integer`
- **Required**: true
- **Min**: `0`

Number of tokens in the input prompt.

#### [completionTokens](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-usage-completiontokens)

- **Path**: `usage.completionTokens`
- **Type**: `integer`
- **Required**: true
- **Min**: `0`

Number of tokens generated in the response.

#### [totalTokens](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-usage-totaltokens)

- **Path**: `usage.totalTokens`
- **Type**: `integer`
- **Required**: true
- **Min**: `0`

Total number of tokens used (prompt + completion).

#### [thinkingTokens](https://runware.ai/docs/models/google-gemini-3-1-flash-lite#response-usage-thinkingtokens)

- **Path**: `usage.thinkingTokens`
- **Type**: `integer`
- **Min**: `0`

Number of tokens used for internal reasoning. Billed separately.

## Examples

### audio-driven (audio-driven)

```
Sure! Here is the action brief based on the voicemail provided.

### 1. Verbatim Transcript
"Hi Mara, this is Celeste from Northline Outfitters following up after our line review. Please reserve 48 undyed canvas chore jackets, style CJ-204: 12 small, 18 medium, 12 large, and 6 extra-large. We also want 36 ribbed merino crews in moss, style MK-17, split evenly across small through extra-large. Target landed cost is no more than $62 for the jacket and 48 for the knit. We need samples by September 18th, 2026. Final confirmation by September 22nd, and delivery to our Portland distribution center between November 4th and 8th. Terms are net 30. Please send the revised line sheet and recycled content certificate. Hold off on the rust colorway. Confirm everything by 3:00 PM Pacific tomorrow. Thanks."

### 2. Summary of Requests
* The client would like to place an order for 48 units of CJ-204 canvas chore jackets and 36 units of MK-17 ribbed merino crews.
* Samples are due September 18th, 2026, with final confirmation by September 22nd. 
* Delivery to the client's Portland distribution center is requested for November 4th-8th.
* The client requested a revised line sheet, recycled content certificate, and to put the rust colorway on hold.

### 3. Order Details
| Style | Color | Total Quantity | Size Breakdown | Max Landed Cost |
| :--- | :--- | :--- | :--- | :--- |
| CJ-204 | Undyed Canvas | 48 | 12S, 18M, 12L, 6XL | $62 |
| MK-17 | Moss | 36 | 9S, 9M, 9L, 9XL | $48 |

### 4. Dates, Terms, and Documentation
* **Sample delivery date:** September 18th, 2026
* **Final confirmation deadline:** September 22nd, 2026
* **Delivery window:** November 4th-8th, 2026
* **Payment terms:** Net 30
* **Required documents:** Revised line sheet and recycled content certificate
* **Hold instruction:** Rust colorway
* **Response deadline:** 3:00 PM Pacific, tomorrow 

### 5. Prioritized Action Items
| Action Item | Owner (Suggested Role) |
| :--- | :--- |
| Acknowledge voicemail and confirm all items by 3 PM tomorrow | Account Manager |
| Coordinate and prepare sample delivery for Sept 18 | Production/Operations |
| Confirm cost requirements are met for landing | Logistics |
| Provide revised line sheet and recycled content certificate | Account Manager/Compliance |
| Hold production for "rust" colorway | Production |

*Note: The voicemail mentions "split evenly across small through extra-large" for the MK-17. 36 items split evenly across 4 sizes equals 9 units per size. This has been noted in the table above.*
```

Finish reason: `stop`

```json
{
  "taskType": "textInference",
  "taskUUID": "edbb355f-654e-4c6b-a284-1f3803a5b947",
  "model": "google:gemini@3.1-flash-lite",
  "inputs": {
    "audios": [
      "https://assets.runware.ai/assets/inputs/44b59a67-2fa0-40f1-89d8-6dc5b2733dad.mp3"
    ]
  },
  "messages": [
    {
      "role": "user",
      "content": "Process this wholesale buyer voicemail into a sales-team action brief. Provide: 1) a faithful verbatim transcript; 2) a concise summary of no more than four bullets; 3) a table listing each style, color, total quantity, size breakdown, and maximum target landed cost; 4) every date, delivery window, payment term, requested document, hold instruction, and response deadline; and 5) prioritized action items with owners suggested by role, such as account manager, production, logistics, or compliance. Clearly flag anything unclear or missing. Do not infer unsupported details or silently correct the speaker."
    }
  ]
}
```

---

### structured-output (structured-output)

```
{
  "lease_date": "2026-02-12",
  "parties": {
    "landlord": "Harbor Point Retail LLC",
    "tenant": "Juniper Bicycle Works, Inc."
  },
  "premises": {
    "suite": "140",
    "street_address": "88 Meridian Avenue",
    "city": "Portland",
    "state": "Maine",
    "rentable_square_feet": 2850,
    "measurement_qualified": true
  },
  "term": {
    "commencement_trigger": "Later of June 1, 2026 or date Landlord delivers premises with base building work substantially complete",
    "fixed_commencement_date": null,
    "term_years": 7,
    "expiration_basis": "Seven years after the Rent Commencement Date",
    "free_rent_months": 3
  },
  "base_rent_schedule": [
    {
      "start_lease_year": 1,
      "end_lease_year": 2,
      "annual_rent_usd": 102600.0,
      "monthly_rent_usd": 8550.0
    },
    {
      "start_lease_year": 3,
      "end_lease_year": 4,
      "annual_rent_usd": 107730.0,
      "monthly_rent_usd": 8977.5
    },
    {
      "start_lease_year": 5,
      "end_lease_year": 6,
      "annual_rent_usd": 113116.5,
      "monthly_rent_usd": 9426.38
    },
    {
      "start_lease_year": 7,
      "end_lease_year": 7,
      "annual_rent_usd": 118772.33,
      "monthly_rent_usd": 9897.69
    }
  ],
  "additional_economics": {
    "operating_expense_share_percent": 4.75,
    "separately_metered_utilities": [
      "electricity",
      "water"
    ],
    "security_deposit_usd": 25650.0
  },
  "renewal_option": {
    "option_count": 1,
    "option_term_years": 5,
    "rent_basis": "Fair market rent, but not less than the base rent payable during the final lease year",
    "notice_earliest_months_before_expiration": 15,
    "notice_latest_months_before_expiration": 12
  },
  "operational_terms": {
    "permitted_use": "Retail sale and repair of bicycles, cycling apparel, and related accessories, including customer test rides in areas designated by Landlord",
    "assignment_consent_standard": "Not to be unreasonably withheld, conditioned, or delayed",
    "landlord_repair_responsibilities": [
      "structural walls",
      "roof membrane",
      "foundation"
    ],
    "liability_insurance_per_occurrence_usd": 2000000.0,
    "liability_insurance_aggregate_usd": 4000000.0
  },
  "review_flags": [
    {
      "field": "security_deposit_usd",
      "severity": "medium",
      "issue": "Discrepancy in security deposit amount between main lease text and Exhibit C",
      "source_values": [
        "25650.0",
        "24000.0"
      ]
    }
  ]
}
```

Finish reason: `stop`

```json
{
  "taskType": "textInference",
  "taskUUID": "74a38d59-5482-4518-a718-a348c3bc354d",
  "outputFormat": "JSON",
  "model": "google:gemini@3.1-flash-lite",
  "messages": [
    {
      "role": "user",
      "content": "Extract the lease terms below into the supplied JSON schema. Use ISO 8601 dates. Express monetary amounts in USD and preserve any unresolved conflict or missing term in review_flags rather than guessing.\n\nLEASE EXCERPT\nLease dated February 12, 2026, between Harbor Point Retail LLC (Landlord) and Juniper Bicycle Works, Inc. (Tenant) for Suite 140, 88 Meridian Avenue, Portland, Maine. The premises contain approximately 2,850 rentable square feet.\n\nThe term begins on the later of June 1, 2026 or the date Landlord delivers the premises with the base building work substantially complete. The term expires seven years after the Rent Commencement Date. Base rent is abated for the first three full calendar months following the Rent Commencement Date. Thereafter, annual base rent is $102,600 for Lease Years 1–2, $107,730 for Years 3–4, $113,116.50 for Years 5–6, and $118,772.33 for Year 7, payable in equal monthly installments.\n\nTenant's proportionate share of operating expenses is 4.75%. Tenant pays separately metered electricity and water. Landlord remains responsible for structural walls, roof membrane, and foundation, except to the extent damage is caused by Tenant.\n\nTenant has one five-year renewal option at fair market rent, but not less than the base rent payable during the final lease year. Tenant must provide written renewal notice no earlier than 15 months and no later than 12 months before expiration. Assignment requires Landlord's prior written consent, not to be unreasonably withheld, conditioned, or delayed.\n\nThe security deposit is $25,650. Exhibit C instead lists the deposit as $24,000. Tenant must maintain commercial general liability insurance of at least $2,000,000 per occurrence and $4,000,000 aggregate. The permitted use is the retail sale and repair of bicycles, cycling apparel, and related accessories, including customer test rides in areas designated by Landlord."
    }
  ],
  "jsonSchema": {
    "name": "retail_lease_abstract",
    "strict": true,
    "schema": {
      "type": "object",
      "properties": {
        "lease_date": {
          "type": "string",
          "description": "Lease execution date in ISO 8601 format."
        },
        "parties": {
          "type": "object",
          "properties": {
            "landlord": {
              "type": "string"
            },
            "tenant": {
              "type": "string"
            }
          },
          "required": [
            "landlord",
            "tenant"
          ],
          "additionalProperties": false
        },
        "premises": {
          "type": "object",
          "properties": {
            "suite": {
              "type": "string"
            },
            "street_address": {
              "type": "string"
            },
            "city": {
              "type": "string"
            },
            "state": {
              "type": "string"
            },
            "rentable_square_feet": {
              "type": "integer"
            },
            "measurement_qualified": {
              "type": "boolean",
              "description": "True when the stated area is approximate or otherwise qualified."
            }
          },
          "required": [
            "suite",
            "street_address",
            "city",
            "state",
            "rentable_square_feet",
            "measurement_qualified"
          ],
          "additionalProperties": false
        },
        "term": {
          "type": "object",
          "properties": {
            "commencement_trigger": {
              "type": "string"
            },
            "fixed_commencement_date": {
              "type": [
                "string",
                "null"
              ]
            },
            "term_years": {
              "type": "integer"
            },
            "expiration_basis": {
              "type": "string"
            },
            "free_rent_months": {
              "type": "integer"
            }
          },
          "required": [
            "commencement_trigger",
            "fixed_commencement_date",
            "term_years",
            "expiration_basis",
            "free_rent_months"
          ],
          "additionalProperties": false
        },
        "base_rent_schedule": {
          "type": "array",
          "items": {
            "type": "object",
            "properties": {
              "start_lease_year": {
                "type": "integer"
              },
              "end_lease_year": {
                "type": "integer"
              },
              "annual_rent_usd": {
                "type": "number"
              },
              "monthly_rent_usd": {
                "type": "number"
              }
            },
            "required": [
              "start_lease_year",
              "end_lease_year",
              "annual_rent_usd",
              "monthly_rent_usd"
            ],
            "additionalProperties": false
          }
        },
        "additional_economics": {
          "type": "object",
          "properties": {
            "operating_expense_share_percent": {
              "type": "number"
            },
            "separately_metered_utilities": {
              "type": "array",
              "items": {
                "type": "string"
              }
            },
            "security_deposit_usd": {
              "type": [
                "number",
                "null"
              ]
            }
          },
          "required": [
            "operating_expense_share_percent",
            "separately_metered_utilities",
            "security_deposit_usd"
          ],
          "additionalProperties": false
        },
        "renewal_option": {
          "type": "object",
          "properties": {
            "option_count": {
              "type": "integer"
            },
            "option_term_years": {
              "type": "integer"
            },
            "rent_basis": {
              "type": "string"
            },
            "notice_earliest_months_before_expiration": {
              "type": "integer"
            },
            "notice_latest_months_before_expiration": {
              "type": "integer"
            }
          },
          "required": [
            "option_count",
            "option_term_years",
            "rent_basis",
            "notice_earliest_months_before_expiration",
            "notice_latest_months_before_expiration"
          ],
          "additionalProperties": false
        },
        "operational_terms": {
          "type": "object",
          "properties": {
            "permitted_use": {
              "type": "string"
            },
            "assignment_consent_standard": {
              "type": "string"
            },
            "landlord_repair_responsibilities": {
              "type": "array",
              "items": {
                "type": "string"
              }
            },
            "liability_insurance_per_occurrence_usd": {
              "type": "number"
            },
            "liability_insurance_aggregate_usd": {
              "type": "number"
            }
          },
          "required": [
            "permitted_use",
            "assignment_consent_standard",
            "landlord_repair_responsibilities",
            "liability_insurance_per_occurrence_usd",
            "liability_insurance_aggregate_usd"
          ],
          "additionalProperties": false
        },
        "review_flags": {
          "type": "array",
          "items": {
            "type": "object",
            "properties": {
              "field": {
                "type": "string"
              },
              "severity": {
                "type": "string",
                "enum": [
                  "low",
                  "medium",
                  "high"
                ]
              },
              "issue": {
                "type": "string"
              },
              "source_values": {
                "type": "array",
                "items": {
                  "type": "string"
                }
              }
            },
            "required": [
              "field",
              "severity",
              "issue",
              "source_values"
            ],
            "additionalProperties": false
          }
        }
      },
      "required": [
        "lease_date",
        "parties",
        "premises",
        "term",
        "base_rent_schedule",
        "additional_economics",
        "renewal_option",
        "operational_terms",
        "review_flags"
      ],
      "additionalProperties": false
    },
    "type": "object",
    "properties": {
      "summary": {
        "type": "string"
      },
      "keyPoints": {
        "type": "array",
        "items": {
          "type": "string"
        }
      },
      "confidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      }
    },
    "required": [
      "summary",
      "keyPoints",
      "confidence"
    ],
    "additionalProperties": false
  }
}
```

---

### tool-calling (tool-calling)

```
{
  "taskType": "textInference",
  "taskUUID": "7052b8c1-9a91-49c9-b689-687e7e84e6b7",
  "cost": 0.000053,
  "finishReason": "tools",
  "tools": [
    {
      "id": "call_117631",
      "name": "get_workspace_status",
      "thoughtSignature": "AY89a1894dJGJ4zVFUkncoE+JH9mVvUvN768wTcwzmFA2t9xZjFwAngJ/uoSTQ7cHkwW0Vii1ZtE/BrL9dGcfVVY2shHv+xGN1T8tQHYhcJhgEI="
    }
  ]
}
```

Finish reason: `tools`

```json
{
  "taskType": "textInference",
  "taskUUID": "7052b8c1-9a91-49c9-b689-687e7e84e6b7",
  "model": "google:gemini@3.1-flash-lite",
  "messages": [
    {
      "role": "user",
      "content": "Check the current workspace status for exhibition project EXH-2026-014, “Threads of the Andes,” before our opening-readiness meeting. Retrieve the latest status of gallery installation, object mounting, lighting, interpretive graphics, lender approvals, security review, and staff training. Include overdue tasks, unresolved blockers, cross-team dependencies, owners, and milestone dates. Use the live workspace rather than relying on prior conversation context; after the status is returned, I’ll need a concise readiness brief for the museum director."
    }
  ],
  "tools": [
    {
      "type": "function",
      "name": "get_workspace_status",
      "description": "Returns the current operational status of the user's workspace.",
      "schema": {}
    }
  ],
  "toolChoice": {
    "type": "tool",
    "name": "get_workspace_status"
  }
}
```

---

### text-generation (text-generation)

```
**Executive Summary:** A shipment of 18 cartons of lot L24-771 experienced a temperature excursion peaking at 11.4°C, resulting in a rejected delivery and necessitating a quality hold.

**Confirmed Facts:** 
* The shipment exceeded the 8°C threshold for approximately 63 minutes (06:18–07:21), with a peak of 11.4°C. 
* The product is currently back at our distribution center at a verified temperature of 6.4°C.
* Packaging remains intact.

**Interpretation:** 
The excursion was likely caused by the courier leaving the van door open during a prior delivery, allowing ambient air infiltration. While current internal temperatures are within range, the product sustained a significant breach of the manufacturer’s 15-minute tolerance window.

**Recommendation:** 
**Quarantine** the shipment. The evidence is insufficient to confirm safety, as the manufacturer has not provided stability data for this duration of exposure.

**Immediate Actions:**
1. **Quality Assurance:** Obtain stability data from the manufacturer for the identified excursion profile. (Lead: QA Manager, Deadline: 11:00 today)
2. **Operations:** Prepare a replacement shipment from existing inventory to be dispatched immediately. (Lead: Logistics Supervisor, Deadline: 11:30 today)
3. **Inventory Management:** Initiate an emergency restock order to replenish regional safety levels. (Lead: Inventory Manager, Deadline: 13:00 today)

**Clinic Communication:**
"We apologize for the delay and are shipping a replacement order immediately to ensure your patients receive their treatment on schedule today."
```

Finish reason: `stop`

```json
{
  "taskType": "textInference",
  "taskUUID": "6840c034-c54a-43bc-94dc-00895814d6a6",
  "model": "google:gemini@3.1-flash-lite",
  "messages": [
    {
      "role": "user",
      "content": "You are supporting operations at a specialty pharmacy distributor. Turn the notes below into a concise incident briefing for the VP of Operations. Use plain natural language and no more than 250 words.\n\nYour briefing must include:\n- a one-sentence executive summary;\n- confirmed facts;\n- the most likely interpretation, clearly distinguished from confirmed facts;\n- a recommendation to release, quarantine, or discard the shipment, with brief reasoning;\n- three immediate actions, each with an owner and deadline;\n- one sentence suitable for sending to the clinic awaiting delivery.\n\nDo not invent regulatory limits or claim the medication is safe. If the available evidence is insufficient for release, say so directly.\n\nIncident notes:\n- Shipment: 18 cartons of refrigerated injectable medication, lot L24-771, intended storage range 2–8°C.\n- Data logger recorded 7.1°C at 06:00, 8.7°C at 06:18, a peak of 11.4°C at 06:42, 9.2°C at 07:03, and 7.8°C at 07:21.\n- Courier arrived at the clinic at 07:28. Clinic staff did not accept the shipment because the logger alarm was active.\n- Cartons were returned to the validated refrigerated van at 07:36 and reached the distribution center at 08:25.\n- Warehouse receiving measured 6.4°C between the center cartons at 08:31. Packaging was intact, with no moisture or visible damage.\n- Courier states the rear door was left open during another delivery from approximately 06:10 to 06:25. Route telemetry shows the refrigeration unit remained powered throughout.\n- The product quality agreement requires manufacturer review for any reading above 8°C lasting more than 15 minutes.\n- Manufacturer stability guidance is not available in the incident file. Quality Assurance has requested it but has not yet received a response.\n- The clinic needs replacement stock by 14:00 today for scheduled patients.\n- Available inventory can cover a replacement shipment, but doing so will reduce regional safety stock from four days to two days."
    }
  ]
}
```

---

### reference-to-text (reference-to-text)

```
### Conformance Report: Prototype Suite vs. Delivered Unit

#### 1. Executive Verdict: Conditional Pass
The delivered unit captures the intended spatial layout and furniture plan, but fails to execute on the critical "warm, high-end" design language defined in the prototype. Significant material omissions—specifically the signature wood-slat headboard wall and the incorrect accent chair color—must be addressed to achieve the brand’s visual identity.

---

#### 2. Prioritized Findings
| Priority | Area | Intended Standard | Observed Deviation | Impact | Recommended Correction | Video Evidence | Confidence |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| **P1** | Wall Treatment | Floor-to-ceiling wood slat feature wall | Painted drywall | Loss of luxury brand identity | Install specified wood slat paneling | 0:02 - 0:04 | High |
| **P2** | FF&E (Accent) | Terra-cotta/orange upholstered chair | Blue upholstered chair | Color scheme dissonance | Replace with correct textile/finish | 0:03 - 0:04 | High |
| **P2** | FF&E (Utility) | Integrated mini-fridge/table unit | Stand-alone white mini-fridge cabinet | Poor aesthetic integration | Install specified integrated casegood | 0:04 - 0:05 | High |
| **P3** | Detailing | Seamless baseboards/transitions | Scuffing and visible caulk gaps | Workmanship/finish quality | Touch up/caulk repair | 0:04 - 0:05 | High |
| **P3** | Decor | Dual framed art pieces | Single small frame | Wall balance/composition | Install set of two as specified | 0:04 | High |

---

#### 3. Cross-Cutting Analysis
The combined effect of these deviations creates a "generic budget" feel that contradicts the "premium boutique" design intent of the prototype. The most striking failure is the lack of the wood-slat feature wall; without it, the lighting design (which is intended to wash the slats) now looks stark against plain white drywall. Furthermore, the selection of a blue accent chair and a white-box mini-fridge appliance introduces "cool" tones that clash with the warm wood tones and beige textiles, undermining the color palette cohesion. The room currently reads as a functional hotel room rather than the curated, atmospheric suite defined in the prototype.

---

#### 4. Pre-Opening Action Plan

**Must Fix Before Handover:**
*   **Casegoods:** Replace the white mini-fridge cabinet with the intended integrated furniture piece to ensure alignment with the room's millwork design.
*   **Wall Art:** Install the correct dual-framed artwork set to anchor the empty space above the utility cabinet.

**Fix Within 30 Days:**
*   **Feature Wall:** Coordinate installation of the wood-slat paneling behind the bed. 
*   **Finishes:** Address all visible wall base damage and caulk imperfections identified at the cabinet transition points.

**Monitor or Verify On Site:**
*   **Accent Furniture:** Audit procurement logs to confirm if the blue chairs are temporary place-holders or a procurement error.
*   **Lighting:** Verify the light temperature (K) of the wall sconces; they appear cooler in the video than the warm, ambient lighting shown in the prototype.

---

#### 5. Verification Checklist (Inspector's Return Visit)
- [ ] **Feature Wall:** Are the vertical wood slats installed, sealed, and properly backlit?
- [ ] **Casegoods:** Is the mini-fridge fully integrated within the casegood console?
- [ ] **Color Palette:** Has the blue accent chair been replaced with the approved terra-cotta/orange model?
- [ ] **Workmanship:** Are the wall-to-baseboard transitions free of gaps and scuffs?
- [ ] **Symmetry:** Is the artwork centered and balanced according to the prototype layout?
```

Finish reason: `stop`

```json
{
  "taskType": "textInference",
  "taskUUID": "c64769af-3de3-474c-a0c5-1ded67f73990",
  "model": "google:gemini@3.1-flash-lite",
  "inputs": {
    "images": [
      "https://assets.runware.ai/assets/inputs/00bc357d-ca99-4d96-948e-776db06e4231.jpg"
    ],
    "videos": [
      "https://assets.runware.ai/assets/inputs/b904e89d-7b6f-4951-80bf-f544e7149e68.mp4"
    ]
  },
  "messages": [
    {
      "role": "user",
      "content": "Act as the hotel operator’s senior design and pre-opening QA inspector. The image is the approved prototype suite and defines the intended design standard; the video is a walkthrough of the delivered room. Analyze them together and produce a decision-ready conformance report, not separate descriptions of each asset.\n\nInfer the prototype’s key design rules—including layout, symmetry, finish quality, color coordination, furniture placement, lighting, and guest circulation—then compare those rules against the installed suite visible throughout the video. Distinguish genuine deviations from uncertainty caused by camera angle, lighting, occlusion, or portions not shown.\n\nReturn a concise professional report in Markdown with:\n1. Executive verdict: Pass, Conditional Pass, or Fail, with a two-sentence rationale.\n2. A prioritized findings table with columns: Priority, Area, Intended Standard, Observed Deviation, Operational or Guest Impact, Recommended Correction, Video Evidence, and Confidence. Use approximate video timestamps or time ranges for evidence.\n3. Cross-cutting analysis explaining how the combined deviations affect the room’s overall design coherence and guest experience; synthesize patterns rather than repeating table entries.\n4. A pre-opening action plan divided into Must Fix Before Handover, Fix Within 30 Days, and Monitor or Verify On Site.\n5. A short verification checklist for the inspector’s return visit.\n\nUse P1 for safety, accessibility, major functional, or brand-critical issues; P2 for conspicuous design or workmanship failures; and P3 for minor cosmetic items. Do not claim exact dimensions, materials, or code violations unless visually supportable. Mark anything that requires physical measurement or close inspection as Verify On Site."
    }
  ]
}
```