---
title: GPT-5 Nano | Runware Docs
url: https://runware.ai/docs/models/openai-gpt-5-nano
description: Fastest, lowest-cost GPT-5 variant for summarization, classification, and lightweight automation
---
# GPT-5 Nano

![GPT-5 Nano](https://assets.runware.ai/covers/openai-gpt-5-nano.jpg)

GPT-5 Nano is OpenAI's fastest and cheapest GPT-5 variant, built for speed- and cost-sensitive workloads such as summarization, classification, extraction, and lightweight automation. It supports 400K context, 128K max output tokens, image input, reasoning token support, structured outputs, and function calling in a compact package optimized for high-throughput API use.

- **Status**: deprecated
- **ID**: `openai-gpt-5-nano`
- Try in Playground
- **Context window**: 400K context
- **Max output**: 128K output
- **Creator**: OpenAI
- **Release Date**: August 7, 2025
- **Capabilities**: Text to Text, Image to Text, Checkpoint

## Pricing

~$0.00062 typical run

**Rates** (per 1M tokens)

- Input: `$0.05`
- Output: `$0.40`
- Cached input: `$0.005`

**4 example generations**

- [tool-calling](https://runware.ai/docs/models/openai-gpt-5-nano/examples#example-live-incident-status-update): `$0.000022` (~1s)
- [reference-to-text](https://runware.ai/docs/models/openai-gpt-5-nano/examples#example-grocery-shelf-availability-audit): `$0.000503` (~10s)
- [text-generation](https://runware.ai/docs/models/openai-gpt-5-nano/examples#example-mobile-dental-clinic-grant-summary): `$0.000737` (~11s)
- [structured-output](https://runware.ai/docs/models/openai-gpt-5-nano/examples#example-retail-lease-risk-register): `$0.00331` (~43s)

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/openai-gpt-5-nano#request-tasktype)

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

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [includeUsage](https://runware.ai/docs/models/openai-gpt-5-nano#request-includeusage)

- **Type**: `boolean`

Include token usage statistics in the response.

### [numberResults](https://runware.ai/docs/models/openai-gpt-5-nano#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.

### [images](https://runware.ai/docs/models/openai-gpt-5-nano#request-inputs-images)

- **Path**: `inputs.images`
- **Type**: `array of strings`
- **Min items**: `1`
- **File size (each)**: `up to 19.07 MB`

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

**Core Parameters**

Primary parameters that define the task output.

### [model](https://runware.ai/docs/models/openai-gpt-5-nano#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `openai-gpt-5-nano`

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/openai-gpt-5-nano#request-seed)

- **Type**: `integer`
- **Min**: `0`
- **Max**: `2147483647`

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/openai-gpt-5-nano#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/openai-gpt-5-nano#request-messages)

- **Path**: `messages.role`
- **Type**: `array of objects (2 properties)`
- **Required**: true
- **Min items**: `1`
- **Max items**: `1000`

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

#### [role](https://runware.ai/docs/models/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#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.

### [maxTokens](https://runware.ai/docs/models/openai-gpt-5-nano#request-settings-maxtokens)

- **Path**: `settings.maxTokens`
- **Type**: `integer`
- **Min**: `1`
- **Max**: `128000`
- **Default**: `32000`

Maximum number of tokens to generate in the response.

### [thinkingLevel](https://runware.ai/docs/models/openai-gpt-5-nano#request-settings-thinkinglevel)

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

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

**Allowed values**: `none` `low` `medium` `high`

### [toolChoice](https://runware.ai/docs/models/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#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`

#### [name](https://runware.ai/docs/models/openai-gpt-5-nano#request-tools-name)

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

Unique function name. Required for function tools.

#### [description](https://runware.ai/docs/models/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#response-tasktype)

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

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/openai-gpt-5-nano#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/openai-gpt-5-nano#response-text)

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

Generated text content.

### [cost](https://runware.ai/docs/models/openai-gpt-5-nano#response-cost)

- **Type**: `float`

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

### [finishReason](https://runware.ai/docs/models/openai-gpt-5-nano#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/openai-gpt-5-nano#response-usage)

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

Token usage statistics for the request.

#### [promptTokens](https://runware.ai/docs/models/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#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/openai-gpt-5-nano#response-usage-thinkingtokens)

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

Number of tokens used for internal reasoning. Billed separately.

## Examples

### structured-output (structured-output)

```
{"summary":"Decision-ready lease abstract and risk register for Harbor Bowl LLC. Premises: Suite 140 (~2,850 rentable square feet) in Meridian Commons, Austin, TX. Term: the Term begins on the earlier of Tenant opening for business or 120 days after Landlord delivers possession, and expires ten years after the Rent Commencement Date; Tenant has two five-year extension options by giving notice within a defined window. Rent: Year 1 Base Rent is $128,250 per year, payable monthly, with a 3% increase on each anniversary of the Rent Commencement Date; Percentage Rent: 6% of annual Gross Sales exceeding the natural breakpoint. Gross Sales includes all sales originating from the Premises, including delivery-platform and online orders fulfilled there, without deduction for third-party commissions. Additional Rent: Tenant pays Proportionate Share (8.4%) of Common Area Costs, property taxes, and insurance; CAM includes management fees not exceeding 5% of total CAM costs; there is no annual cap on CAM/other costs; monthly estimates and annual reconciliation with a 20-day payment window; audit rights once per calendar year. Delivery/Improvements: Landlord delivers broom-clean with existing utility stubs; Tenant responsible for design, permitting, grease interceptor work, utility upgrades, and construction; TI allowance of $45 per RSF reimbursed after opening with lien waivers and no-default; any unused allowance forfeited; Landlord makes no representation on capacity of electrical, gas, water, sewer, or exhaust systems. Use: Permitted Use is a fast-casual restaurant (grain bowls, salads, nonalcoholic beverages) with continuous operation seven days per week during center hours; failure to operate for more than five consecutive days permits Landlord to collect $500 per day as additional rent; no exclusive-use protection. Repairs/Compliance: Tenant maintains and replaces all equipment and systems serving the Premises; Landlord maintains roof, foundation, and structural walls with cost included in CAM unless Tenant-caused. Assignment/Transfer: Tenant may not assign or sublet or transfer >25% ownership without Landlord’s consent (sole discretion); Landlord may recapture the Premises. Default: monetary defaults cureable in five days after notice; nonmonetary defaults cureable in ten days, with extensions if cannot be cured promptly; Landlord may issue up to two monetary-default notices in any 12-month period. Indemnity/Insurance: Tenant indemnifies Landlord for claims arising on the Premises, with exceptions for Landlord’s gross negligence or willful misconduct; minimum $5,000,000 per occurrence GL, 18 months of business interruption coverage, and other coverages as Landlord reasonably requires; Landlord not liable for lost profits. Guaranty: Harbor Bowl Holdings Inc. to execute Landlord’s standard continuing guaranty covering all Tenant obligations, not reduced during term and surviving assignment. Relocation: After Year 5, Landlord may relocate Tenant to a comparable-premises center; Landlord pays moving expenses and unamortized initial improvements; no termination right for relocation. Casualty: Landlord may restore or terminate after casualty; Base Rent abates proportionately while Premises is unusable, but Additional Rent continues; Tenant may terminate only if restoration is not substantially complete within 270 days. Tenant-provided business assumptions note planned opening ~180 days after possession, estimated build-out cost ~$650,000, projected first-year Gross Sales ~$2,400,000, delivery-platform commissions ~24% of delivery sales, uninterrupted operation is important to catering contracts, and parent company will not approve an unlimited guaranty.","keyPoints":["Risk (open question): Rent Commencement Date and timing ambiguity between Term start and Rent escalations. Section references: Term starts at the earlier of opening or 120 days after possession, and Rent escalates on anniversaries of the Rent Commencement Date. Quote: 'The Term begins on the earlier of (a) Tenant opening for business or (b) 120 days after Landlord delivers possession' and 'Base Rent increases by 3% on each anniversary of the Rent Commencement Date.'","Risk (open question): Definition and computation of the natural breakpoint for Percentage Rent; potential for growth in rent based on sales. Section references: §3 Base Rent and Percentage Rent. Quotes: 'Percentage Rent equal to 6% of annual Gross Sales exceeding the natural breakpoint' and 'Gross Sales includes all sales originating from the Premises, including delivery-platform and online orders fulfilled there, without deduction for third-party commissions.'","Risk: Absence of an annual cap on Common Area Costs; potential for escalating occupancy costs. Section: §4 Additional Rent. Quote: 'There is no annual cap.'","Risk: CAM charge composition includes potential increases from management fees and capital expenditures; pass-through mechanics. Section: §4. Quotes: 'Common Area Costs include management fees not exceeding 5% of total Common Area Costs and capital expenditures amortized over the useful life determined by Landlord.'","Risk: TI allowance is finite and forfeited if unused; no guarantees about utility capacity. Sections: §6 Delivery and Improvements. Quotes: 'Landlord provides a tenant-improvement allowance of $45 per rentable square foot, reimbursed after Tenant opens... Any unused allowance is forfeited.' and 'Landlord makes no representation regarding the capacity of electrical, gas, water, sewer, or exhaust systems.'","Risk: Tenant is responsible for all design, permitting, grease interceptor, utility upgrades, and construction, potentially increasing upfront costs. Section: §6 Delivery and Improvements. Quote: 'Tenant is responsible for all design, permitting, grease interceptor work, utility upgrades, and construction.'","Risk: Operational covenant risk and cost for failure to operate; potential daily charge. Section: §7 Use and Operation. Quote: 'Tenant must operate continuously seven days per week during center hours' and 'Failure to operate for more than five consecutive days permits Landlord to collect, as additional rent, $500 per day until operations resume.'","Risk: Exclusive-use protection is not granted; potential competitive exposure. Section: §7 Use and Operation. Quote: 'Tenant receives no exclusive-use protection.'","Risk: Assignment and transfer restrictions; Landlord discretion and recapture risk. Section: §12 Assignment. Quotes: 'Tenant may not assign... without Landlord's prior written consent, which may be withheld in Landlord's sole discretion' and 'Landlord may recapture the Premises instead of approving a proposed assignment or sublease.'","Risk: Short monetary cure period and cure mechanics; risk of eviction for defaults. Section: §15 Default. Quotes: 'Monetary defaults have a five-day cure period after written notice' and 'Notwithstanding the foregoing, Landlord is not required to provide more than two notices of monetary default in any 12-month period.'","Risk: Guaranty terms create ongoing liability; parent company decline of unlimited guaranty is an open issue. Section: §21 Guaranty. Quotes: 'Harbor Bowl Holdings Inc. shall execute Landlord's standard form of continuing guaranty' and 'The guaranty does not burn off or reduce during the Term and remains effective after any assignment.'","Risk: Relocation risk post Year 5; no termination right allowed. Section: §24 Relocation. Quote: 'After the fifth Lease Year, Landlord may relocate Tenant to other premises in the center of comparable size' and 'Tenant has no termination right in connection with relocation.'","Risk: Casualty economics: rent abatement during casualty but continued Additional Rent; restoration timeline cap. Section: §27 Casualty. Quote: 'Base Rent abates proportionately while the Premises is unusable, but Additional Rent continues' and 'Tenant may terminate only if restoration is not substantially complete within 270 days after the casualty.'","Open question: Alignment between Tenant’s stated business assumptions (opening ~180 days after possession) and the Lease’s Term mechanics (opening or 120 days after possession)."],"category":"Lease Abstract and Risk Register"}
```

Finish reason: `stop`

```json
{
  "taskType": "textInference",
  "taskUUID": "9f51ac06-99d8-4297-a48a-f37be4337ef2",
  "outputFormat": "JSON",
  "model": "openai:gpt@5-nano",
  "settings": {
    "thinkingLevel": "medium"
  },
  "messages": [
    {
      "role": "user",
      "content": "Analyze the following proposed retail lease for Harbor Bowl LLC, a fast-casual restaurant operator. Produce a decision-ready lease abstract and risk register for the operator's real-estate counsel and finance team. Use only the supplied text. Cite the relevant section and quote short supporting language for every risk. Do not treat missing information as fact; identify it as an open question. Return only JSON matching the supplied schema.\n\nPROPOSED LEASE EXCERPT\n\nPremises: Suite 140, approximately 2,850 rentable square feet, Meridian Commons, Austin, Texas.\n\n§2 Term. The Term begins on the earlier of (a) Tenant opening for business or (b) 120 days after Landlord delivers possession. The Term expires ten years after the Rent Commencement Date. Tenant has two five-year extension options by giving notice not less than 12 months and not more than 15 months before the then-current expiration date. No option may be exercised if Tenant is then in default beyond any applicable cure period.\n\n§3 Base Rent. Year 1 Base Rent is $128,250 per year, payable monthly. Base Rent increases by 3% on each anniversary of the Rent Commencement Date. Tenant shall also pay Percentage Rent equal to 6% of annual Gross Sales exceeding the natural breakpoint. Gross Sales includes all sales originating from the Premises, including delivery-platform and online orders fulfilled there, without deduction for third-party commissions.\n\n§4 Additional Rent. Tenant shall pay its Proportionate Share, stated as 8.4%, of Common Area Costs, property taxes, and insurance. Common Area Costs include management fees not exceeding 5% of total Common Area Costs and capital expenditures amortized over the useful life determined by Landlord. There is no annual cap. Tenant shall pay estimates monthly and any reconciliation balance within 20 days after notice. Tenant may audit records once per calendar year by notice given within 60 days after receiving the reconciliation statement.\n\n§6 Delivery and Improvements. Landlord will deliver the Premises broom-clean with existing utility stubs in their current condition. Tenant accepts the Premises as-is. Tenant is responsible for all design, permitting, grease interceptor work, utility upgrades, and construction. Landlord provides a tenant-improvement allowance of $45 per rentable square foot, reimbursed after Tenant opens, provides final lien waivers, and demonstrates that no default exists. Any unused allowance is forfeited. Landlord makes no representation regarding the capacity of electrical, gas, water, sewer, or exhaust systems.\n\n§7 Use and Operation. Permitted Use is a fast-casual restaurant specializing in grain bowls, salads, and nonalcoholic beverages, and no other use. Tenant must operate continuously seven days per week during center hours, except for casualty, force majeure, or repairs approved by Landlord. Failure to operate for more than five consecutive days permits Landlord to collect, as additional rent, $500 per day until operations resume. Tenant receives no exclusive-use protection.\n\n§9 Repairs and Compliance. Tenant shall maintain and replace all equipment and systems exclusively serving the Premises, including HVAC, plumbing lines, grease interceptor, fire suppression, and exhaust equipment, regardless of whether installed by Tenant. Tenant shall comply with all laws affecting Tenant's use or occupancy, including accessibility requirements. Landlord maintains the roof, foundation, and structural walls, with the cost included in Common Area Costs unless caused by Tenant.\n\n§12 Assignment. Tenant may not assign the Lease, sublet any portion of the Premises, or undergo a transfer of more than 25% of its ownership interests without Landlord's prior written consent, which may be withheld in Landlord's sole discretion. Any approved transfer does not release Tenant or any guarantor. Landlord may recapture the Premises instead of approving a proposed assignment or sublease.\n\n§15 Default. Monetary defaults have a five-day cure period after written notice. Nonmonetary defaults have a ten-day cure period, provided that if a default cannot reasonably be cured within ten days, Tenant must begin cure within that period and diligently complete it. Notwithstanding the foregoing, Landlord is not required to provide more than two notices of monetary default in any 12-month period.\n\n§18 Indemnity and Insurance. Tenant indemnifies Landlord against all claims arising in or about the Premises, except to the extent caused by Landlord's gross negligence or willful misconduct. Tenant must maintain commercial general liability insurance of at least $5,000,000 per occurrence, business interruption coverage for 18 months, and any other coverage Landlord reasonably requires. Landlord is not liable for lost profits or interruption of Tenant's business.\n\n§21 Guaranty. Harbor Bowl Holdings Inc. shall execute Landlord's standard form of continuing guaranty covering all Tenant obligations. The guaranty does not burn off or reduce during the Term and remains effective after any assignment.\n\n§24 Relocation. After the fifth Lease Year, Landlord may relocate Tenant to other premises in the center of comparable size. Landlord will pay reasonable moving expenses and the unamortized cost of Tenant's initial improvements, calculated on a straight-line basis over ten years. Tenant has no termination right in connection with relocation.\n\n§27 Casualty. If the Premises is materially damaged, Landlord may elect to restore or terminate the Lease. Base Rent abates proportionately while the Premises is unusable, but Additional Rent continues. Tenant may terminate only if restoration is not substantially complete within 270 days after the casualty.\n\nBusiness assumptions supplied by Tenant: planned opening 180 days after possession; estimated initial build-out cost of $650,000; projected first-year Gross Sales of $2,400,000; delivery-platform commissions average 24% of delivery sales; uninterrupted operation at this location is important to catering contracts; the parent company will not approve an unlimited guaranty."
    }
  ],
  "jsonSchema": {
    "name": "response",
    "strict": true,
    "schema": {
      "type": "object",
      "properties": {
        "summary": {
          "type": "string"
        },
        "keyPoints": {
          "type": "array",
          "items": {
            "type": "string"
          }
        },
        "category": {
          "type": "string"
        }
      },
      "required": [
        "summary",
        "keyPoints",
        "category"
      ],
      "additionalProperties": false
    }
  }
}
```

---

### text-generation (text-generation)

```
Executive Summary: Expanding Riverbend’s Mobile Dental Clinic to Improve Access and Outcomes in Rural Communities

Riverbend Community Health, a nonprofit federally qualified health center serving three rural counties, requests $185,000 to expand its mobile dental clinic for a 12-month period. The expansion responds to persistent access barriers identified by our patients: transportation challenges and long wait times, including an average nine-week wait at our fixed dental clinic. In the most recent program year, the mobile unit delivered 1,840 patient visits with 71% of patients uninsured or covered by Medicaid. This proposed investment strengthens care delivery where it is most needed, directly addressing inequities in rural dental access.

The expansion adds two weekly service days and four new stops—two elementary schools, a food pantry, and a farmworker resource center—to extend reach into communities with the greatest transportation and access barriers. Grant funds will support a 0.8 FTE dental hygienist, a part-time driver and patient navigator, portable X-ray equipment, clinical supplies, fuel, and interpretation services in Spanish and Mixteco. These investments enable a more reliable, culturally responsive mobile service that can operate closer to where patients live, work, and gather.

During the grant period, Riverbend plans to deliver 1,200 additional visits to at least 750 unique patients. Services will encompass screenings, cleanings, fluoride treatments, sealants, X-rays, simple restorations, referrals, and assistance enrolling eligible patients in Medicaid. The unit will track visits, unique patients, completed treatment plans, referral completion, insurance enrollment, patient demographics, and satisfaction, facilitating rigorous measurement of outcomes. The target is for 65% of patients with identified treatment needs to complete a treatment plan or specialist referral within six months, a measurable standard that balances timeliness with the realities of care coordination in rural settings.

This proposal aligns with the Brightwell County Health Equity Fund’s emphasis on improved access, measurable outcomes, culturally responsive care, and sustainable programs. Riverbend anticipates that ongoing costs will be supported through Medicaid reimbursement, sliding-fee revenue, and the integration of the expanded mobile route into our operating budget. To ensure continuity of care beyond the mobile service, the program includes referral partnerships with two local dental practices for cases beyond the mobile unit’s scope. By expanding access and delivering timely, appropriate care, Riverbend will reduce barriers to treatment, improve oral health outcomes, and advance health equity for rural families across three counties.
```

Finish reason: `stop`

```json
{
  "taskType": "textInference",
  "taskUUID": "86ed748e-0b76-4695-807e-b68909d77b27",
  "model": "openai:gpt@5-nano",
  "settings": {
    "thinkingLevel": "medium",
    "maxTokens": 2000
  },
  "messages": [
    {
      "role": "user",
      "content": "Draft a persuasive 400–500 word executive summary for Riverbend Community Health’s grant application to the Brightwell County Health Equity Fund. The request is $185,000 for a 12-month expansion of its mobile dental clinic.\n\nUse only the facts below:\n\nRiverbend Community Health is a nonprofit federally qualified health center serving three rural counties. Its mobile dental unit currently visits six elementary schools and two senior centers. During the last completed program year, the unit provided 1,840 patient visits; 71% of patients were uninsured or covered by Medicaid. Average appointment wait time at Riverbend’s fixed dental clinic is nine weeks. Transportation is the most frequently reported barrier to care in Riverbend’s annual patient survey.\n\nThe proposed expansion will add two weekly service days and four new stops: two schools, a food pantry, and a farmworker resource center. Grant funds will support a 0.8 FTE dental hygienist, a part-time driver and patient navigator, portable X-ray equipment, clinical supplies, fuel, and interpretation services in Spanish and Mixteco.\n\nDuring the grant period, Riverbend plans to deliver 1,200 additional visits to at least 750 unique patients. Services will include screenings, cleanings, fluoride treatments, sealants, X-rays, simple restorations, referrals, and help enrolling eligible patients in Medicaid. Riverbend will track visits, unique patients, completed treatment plans, referral completion, insurance enrollment, patient demographics, and satisfaction. Its target is for 65% of patients with identified treatment needs to complete a treatment plan or specialist referral within six months.\n\nThe fund prioritizes improved access, measurable outcomes, culturally responsive care, and plans that can continue after grant funding. Riverbend expects ongoing costs to be supported through Medicaid reimbursement, sliding-fee revenue, and integration of the expanded route into its operating budget. It also plans referral partnerships with two local dental practices for cases beyond the mobile unit’s scope.\n\nWrite for a foundation review committee. Open with the community need and funding request, connect the activities directly to measurable outcomes, and close with the sustainability case. Keep the tone credible, specific, and urgent without exaggeration. Use a short heading followed by cohesive prose, not bullets. Do not invent quotations, statistics, partners, awards, or patient stories. Return only the finished executive summary."
    }
  ]
}
```

---

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

```
Field Merchandising Audit – Retail Shelf (Beverages)

Date of observation: [today]

Overview of the display
- Multi-shelf beverage section with primarily aluminum cans and a few bottle packs. Colors are dominated by blue/silver can designs with some darker and light green variants interspersed.
- The display appears full across visible shelves; no large open gaps or empty product bays are evident at first glance.

Stock status and shelf location (stockouts/low stock)
- No confirmed stockouts observed on any shelf row shown. All visible bays appear stocked with products extending to the shelf fronts.
- If any stockouts exist, they are not clearly visible in this image (no empty full-width slots or obvious missing SKUs). Note: adjacent aisle ends or lower levels not captured.

Facings and product variety (approximate)
- Top shelf: Predominantly tall cans with many facing the front; uniform can height makes exact facings hard to count, but product appears evenly displayed with front-facing products across most positions.
- Middle shelves: A mix of blue/silver can variants; facing appears generally forward with slight rotation in a few places (see abnormalities below).
- Lower shelves: A mix of larger can packs and some blue-toned 12-packs; front facings are generally aligned, with a few instances where items are angled or pushed back from the shelf edge.
- Distinguishable variants: Several color tones (blue, silver, green) indicate multiple SKUs or brands, but exact SKUs/brands are not legible enough to distinguish precisely. If labels are unreadable, note accordingly.

Misplaced, rotated, damaged, or poorly aligned items
- Some items on the middle-right area appear slightly angled or not perfectly flush to the front edge; these are minor and do not indicate widespread misalignment.
- A few cans near the bottom shelf toward the left appear rotated or not fully facing the shopper; still within acceptable presentation but worth attention.
- No obvious damaged packaging visible; no items hanging off shelves or collapsed packaging noted.

Shelf-edge prices and promotional tags
- Shelf-edge price labels are present along the front of each shelf. Most appear legible at a distance but finer details (exact prices) are not clearly readable in this image.
- A prominent promotional tag or sale sign is visible on the middle-left lower shelf (yellow tag) with a price of 29.99 on a central block; its exact product alignment relative to the rest of the row is not fully clear but it’s clearly displayed.
- Some small yellow/orange tag notes are present on the left side of several shelves, but unreadable in this image due to resolution.

Cleanliness and maintenance
- Shelves appear clean with no obvious grime or spillage on the shelf lips.
- No visible dust accumulation; packaging looks intact.
- No clutter in the aisles shown; shelf boundaries appear orderly.

Uncertain or unreadable elements
- Exact product SKUs, brands, and variant names on each can are not legible from the image. Where unreadable, note as unreadable rather than guessing.
- Some price tag details are not legible; only the clearly visible 29.99 promotional tag is confirmed.
- Planogram compliance cannot be assessed due to lack of reference planogram.

Prioritized corrective actions
1. Verify and label readability: If feasible, ensure all shelf-edge price labels and promotional tags are clearly legible from shopper distance; consider higher-contrast tags for key SKUs.
2. Tighten facings on identified spots: In the mid-to-lower shelves where rotation or backward pushing is observed, adjust to full front-facing alignment to maintain a clean, uniform appearance.
3. Routine cleanliness check: Continue daily wipe-downs of shelf lips and front edges to maintain a pristine look; this is already adequate but ongoing maintenance is recommended.
4. Stock verification: If store operations permit, conduct a quick stocking check to confirm no hidden stockouts behind front-facing items, especially in the middle rows where variation exists.
5. Readability improvement plan: If possible, refresh or replace unreadable price/brand labels to support quicker shopper scanning.

Notes
- This audit reflects only what is visibly supported in the attached image.
- No claims of planogram compliance or noncompliance are made due to missing reference planogram.

If you’d like, I can annotate a version with explicit sections for each shelf level (top/middle/lower) and mark specific coordinates where rotation or tag visibility issues were observed.
```

Finish reason: `stop`

```json
{
  "taskType": "textInference",
  "taskUUID": "f3b80ab0-b7fe-48cd-bd0a-b13d6225a723",
  "model": "openai:gpt@5-nano",
  "settings": {
    "thinkingLevel": "low",
    "maxTokens": 1200
  },
  "inputs": {
    "images": [
      "https://assets.runware.ai/assets/inputs/93fa934f-d591-4226-b030-c24cbf52b936.jpg"
    ]
  },
  "messages": [
    {
      "role": "user",
      "content": "Inspect the attached retail-shelf photo and write a concise field merchandising audit in plain text with clear headings. Document only what is visibly supported. Include: an overview of the display; observable stockouts or low-stock areas with shelf location; approximate facings by visible product or variant where distinguishable; misplaced, rotated, damaged, or poorly aligned items; visible shelf-edge prices and promotional tags; cleanliness or maintenance issues; and prioritized corrective actions. Distinguish confirmed observations from anything uncertain. If a label, price, brand, or product variant is unreadable, say so rather than guessing. Do not claim planogram noncompliance because no reference planogram is provided."
    }
  ]
}
```

---

### tool-calling (tool-calling)

```
{
  "taskType": "textInference",
  "taskUUID": "541cdcc7-a83a-4d9c-8b4c-97318be592dc",
  "cost": 0.000022,
  "finishReason": "tools",
  "tools": [
    {
      "id": "call_vkIYn7HL5Xl4kwlTjSaz5hgM",
      "name": "get_current_utc_time"
    }
  ]
}
```

Finish reason: `tools`

```json
{
  "taskType": "textInference",
  "taskUUID": "541cdcc7-a83a-4d9c-8b4c-97318be592dc",
  "model": "openai:gpt@5-nano",
  "settings": {
    "thinkingLevel": "low",
    "maxTokens": 800
  },
  "messages": [
    {
      "role": "user",
      "content": "Prepare the final public status-page update for a resolved payment-processing incident. The incident began at 14:20 UTC on August 31, 2026, and service has just been fully restored. First, call the available function to obtain the exact current UTC date and time; do not estimate it or draft the notice before receiving the function result. After the developer executes the call and sends the result in a follow-up turn, write a concise customer-facing update with: a clear “Resolved” heading, the restoration timestamp in UTC, the incident duration calculated from the start time, confirmation that queued transactions are processing normally, and a brief apology. Do not invent a cause."
    }
  ],
  "tools": [
    {
      "type": "function",
      "name": "get_current_utc_time",
      "description": "Returns the current date and time in UTC.",
      "schema": {}
    }
  ],
  "toolChoice": {
    "type": "tool",
    "name": "get_current_utc_time"
  }
}
```