Moderations

Create moderation

ModerationCreateResponse moderations().create(ModerationCreateParamsparams, RequestOptionsrequestOptions = RequestOptions.none())

post /moderations

Classifies if text and/or image inputs are potentially harmful. Learn more in the moderation guide.

Parameters

ModerationCreateParams params
- Input input
  
  Input (or inputs) to classify. Can be a single string, an array of strings, or an array of multi-modal input objects similar to other models.
  - String
  - List<String>
  - List<ModerationMultiModalInput>
    - class ModerationImageUrlInput:
      
      An object describing an image to classify.
      - ImageUrl imageUrl
        
        Contains either an image URL or a data URL for a base64 encoded image.
        
        String url
        
        Either a URL of the image or the base64 encoded image data.
      - JsonValue; type "image_url"constant
        
        Always image_url.
        
        IMAGE_URL("image_url")
    - class ModerationTextInput:
      
      An object describing text to classify.
      - String text
        
        A string of text to classify.
      - JsonValue; type "text"constant
        
        Always text.
        
        TEXT("text")
- Optional<ModerationModel> model
  
  The content moderation model you would like to use. Learn more in the moderation guide, and learn about available models here.

Returns

class ModerationCreateResponse:

Represents if a given text input is potentially harmful.
- String id
  
  The unique identifier for the moderation request.
- String model
  
  The model used to generate the moderation results.
- List<Moderation> results
  
  A list of moderation objects.
  - Categories categories
    
    A list of the categories, and whether they are flagged or not.
    - boolean harassment
      
      Content that expresses, incites, or promotes harassing language towards any target.
    - boolean harassmentThreatening
      
      Harassment content that also includes violence or serious harm towards any target.
    - boolean hate
      
      Content that expresses, incites, or promotes hate based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste. Hateful content aimed at non-protected groups (e.g., chess players) is harassment.
    - boolean hateThreatening
      
      Hateful content that also includes violence or serious harm towards the targeted group based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste.
    - Optional<Boolean> illicit
      
      Content that includes instructions or advice that facilitate the planning or execution of wrongdoing, or that gives advice or instruction on how to commit illicit acts. For example, "how to shoplift" would fit this category.
    - Optional<Boolean> illicitViolent
      
      Content that includes instructions or advice that facilitate the planning or execution of wrongdoing that also includes violence, or that gives advice or instruction on the procurement of any weapon.
    - boolean selfHarm
      
      Content that promotes, encourages, or depicts acts of self-harm, such as suicide, cutting, and eating disorders.
    - boolean selfHarmInstructions
      
      Content that encourages performing acts of self-harm, such as suicide, cutting, and eating disorders, or that gives instructions or advice on how to commit such acts.
    - boolean selfHarmIntent
      
      Content where the speaker expresses that they are engaging or intend to engage in acts of self-harm, such as suicide, cutting, and eating disorders.
    - boolean sexual
      
      Content meant to arouse sexual excitement, such as the description of sexual activity, or that promotes sexual services (excluding sex education and wellness).
    - boolean sexualMinors
      
      Sexual content that includes an individual who is under 18 years old.
    - boolean violence
      
      Content that depicts death, violence, or physical injury.
    - boolean violenceGraphic
      
      Content that depicts death, violence, or physical injury in graphic detail.
  - CategoryAppliedInputTypes categoryAppliedInputTypes
    
    A list of the categories along with the input type(s) that the score applies to.
    - List<Harassment> harassment
      
      The applied input type(s) for the category 'harassment'.
      - TEXT("text")
    - List<HarassmentThreatening> harassmentThreatening
      
      The applied input type(s) for the category 'harassment/threatening'.
      - TEXT("text")
    - List<Hate> hate
      
      The applied input type(s) for the category 'hate'.
      - TEXT("text")
    - List<HateThreatening> hateThreatening
      
      The applied input type(s) for the category 'hate/threatening'.
      - TEXT("text")
    - List<Illicit> illicit
      
      The applied input type(s) for the category 'illicit'.
      - TEXT("text")
    - List<IllicitViolent> illicitViolent
      
      The applied input type(s) for the category 'illicit/violent'.
      - TEXT("text")
    - List<SelfHarm> selfHarm
      
      The applied input type(s) for the category 'self-harm'.
      - TEXT("text")
      - IMAGE("image")
    - List<SelfHarmInstruction> selfHarmInstructions
      
      The applied input type(s) for the category 'self-harm/instructions'.
      - TEXT("text")
      - IMAGE("image")
    - List<SelfHarmIntent> selfHarmIntent
      
      The applied input type(s) for the category 'self-harm/intent'.
      - TEXT("text")
      - IMAGE("image")
    - List<Sexual> sexual
      
      The applied input type(s) for the category 'sexual'.
      - TEXT("text")
      - IMAGE("image")
    - List<SexualMinor> sexualMinors
      
      The applied input type(s) for the category 'sexual/minors'.
      - TEXT("text")
    - List<Violence> violence
      
      The applied input type(s) for the category 'violence'.
      - TEXT("text")
      - IMAGE("image")
    - List<ViolenceGraphic> violenceGraphic
      
      The applied input type(s) for the category 'violence/graphic'.
      - TEXT("text")
      - IMAGE("image")
  - CategoryScores categoryScores
    
    A list of the categories along with their scores as predicted by model.
    - double harassment
      
      The score for the category 'harassment'.
    - double harassmentThreatening
      
      The score for the category 'harassment/threatening'.
    - double hate
      
      The score for the category 'hate'.
    - double hateThreatening
      
      The score for the category 'hate/threatening'.
    - double illicit
      
      The score for the category 'illicit'.
    - double illicitViolent
      
      The score for the category 'illicit/violent'.
    - double selfHarm
      
      The score for the category 'self-harm'.
    - double selfHarmInstructions
      
      The score for the category 'self-harm/instructions'.
    - double selfHarmIntent
      
      The score for the category 'self-harm/intent'.
    - double sexual
      
      The score for the category 'sexual'.
    - double sexualMinors
      
      The score for the category 'sexual/minors'.
    - double violence
      
      The score for the category 'violence'.
    - double violenceGraphic
      
      The score for the category 'violence/graphic'.
  - boolean flagged
    
    Whether any of the below categories are flagged.

Example

package com.openai.example;

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.moderations.ModerationCreateParams;
import com.openai.models.moderations.ModerationCreateResponse;

public final class Main {
    private Main() {}

    public static void main(String[] args) {
        OpenAIClient client = OpenAIOkHttpClient.fromEnv();

        ModerationCreateParams params = ModerationCreateParams.builder()
            .input("I want to kill them.")
            .build();
        ModerationCreateResponse moderation = client.moderations().create(params);
    }
}

Response

{
  "id": "id",
  "model": "model",
  "results": [
    {
      "categories": {
        "harassment": true,
        "harassment/threatening": true,
        "hate": true,
        "hate/threatening": true,
        "illicit": true,
        "illicit/violent": true,
        "self-harm": true,
        "self-harm/instructions": true,
        "self-harm/intent": true,
        "sexual": true,
        "sexual/minors": true,
        "violence": true,
        "violence/graphic": true
      },
      "category_applied_input_types": {
        "harassment": [
          "text"
        ],
        "harassment/threatening": [
          "text"
        ],
        "hate": [
          "text"
        ],
        "hate/threatening": [
          "text"
        ],
        "illicit": [
          "text"
        ],
        "illicit/violent": [
          "text"
        ],
        "self-harm": [
          "text"
        ],
        "self-harm/instructions": [
          "text"
        ],
        "self-harm/intent": [
          "text"
        ],
        "sexual": [
          "text"
        ],
        "sexual/minors": [
          "text"
        ],
        "violence": [
          "text"
        ],
        "violence/graphic": [
          "text"
        ]
      },
      "category_scores": {
        "harassment": 0,
        "harassment/threatening": 0,
        "hate": 0,
        "hate/threatening": 0,
        "illicit": 0,
        "illicit/violent": 0,
        "self-harm": 0,
        "self-harm/instructions": 0,
        "self-harm/intent": 0,
        "sexual": 0,
        "sexual/minors": 0,
        "violence": 0,
        "violence/graphic": 0
      },
      "flagged": true
    }
  ]
}

Domain Types

Moderation

class Moderation:
- Categories categories
  
  A list of the categories, and whether they are flagged or not.
  - boolean harassment
    
    Content that expresses, incites, or promotes harassing language towards any target.
  - boolean harassmentThreatening
    
    Harassment content that also includes violence or serious harm towards any target.
  - boolean hate
    
    Content that expresses, incites, or promotes hate based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste. Hateful content aimed at non-protected groups (e.g., chess players) is harassment.
  - boolean hateThreatening
    
    Hateful content that also includes violence or serious harm towards the targeted group based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste.
  - Optional<Boolean> illicit
    
    Content that includes instructions or advice that facilitate the planning or execution of wrongdoing, or that gives advice or instruction on how to commit illicit acts. For example, "how to shoplift" would fit this category.
  - Optional<Boolean> illicitViolent
    
    Content that includes instructions or advice that facilitate the planning or execution of wrongdoing that also includes violence, or that gives advice or instruction on the procurement of any weapon.
  - boolean selfHarm
    
    Content that promotes, encourages, or depicts acts of self-harm, such as suicide, cutting, and eating disorders.
  - boolean selfHarmInstructions
    
    Content that encourages performing acts of self-harm, such as suicide, cutting, and eating disorders, or that gives instructions or advice on how to commit such acts.
  - boolean selfHarmIntent
    
    Content where the speaker expresses that they are engaging or intend to engage in acts of self-harm, such as suicide, cutting, and eating disorders.
  - boolean sexual
    
    Content meant to arouse sexual excitement, such as the description of sexual activity, or that promotes sexual services (excluding sex education and wellness).
  - boolean sexualMinors
    
    Sexual content that includes an individual who is under 18 years old.
  - boolean violence
    
    Content that depicts death, violence, or physical injury.
  - boolean violenceGraphic
    
    Content that depicts death, violence, or physical injury in graphic detail.
- CategoryAppliedInputTypes categoryAppliedInputTypes
  
  A list of the categories along with the input type(s) that the score applies to.
  - List<Harassment> harassment
    
    The applied input type(s) for the category 'harassment'.
    - TEXT("text")
  - List<HarassmentThreatening> harassmentThreatening
    
    The applied input type(s) for the category 'harassment/threatening'.
    - TEXT("text")
  - List<Hate> hate
    
    The applied input type(s) for the category 'hate'.
    - TEXT("text")
  - List<HateThreatening> hateThreatening
    
    The applied input type(s) for the category 'hate/threatening'.
    - TEXT("text")
  - List<Illicit> illicit
    
    The applied input type(s) for the category 'illicit'.
    - TEXT("text")
  - List<IllicitViolent> illicitViolent
    
    The applied input type(s) for the category 'illicit/violent'.
    - TEXT("text")
  - List<SelfHarm> selfHarm
    
    The applied input type(s) for the category 'self-harm'.
    - TEXT("text")
    - IMAGE("image")
  - List<SelfHarmInstruction> selfHarmInstructions
    
    The applied input type(s) for the category 'self-harm/instructions'.
    - TEXT("text")
    - IMAGE("image")
  - List<SelfHarmIntent> selfHarmIntent
    
    The applied input type(s) for the category 'self-harm/intent'.
    - TEXT("text")
    - IMAGE("image")
  - List<Sexual> sexual
    
    The applied input type(s) for the category 'sexual'.
    - TEXT("text")
    - IMAGE("image")
  - List<SexualMinor> sexualMinors
    
    The applied input type(s) for the category 'sexual/minors'.
    - TEXT("text")
  - List<Violence> violence
    
    The applied input type(s) for the category 'violence'.
    - TEXT("text")
    - IMAGE("image")
  - List<ViolenceGraphic> violenceGraphic
    
    The applied input type(s) for the category 'violence/graphic'.
    - TEXT("text")
    - IMAGE("image")
- CategoryScores categoryScores
  
  A list of the categories along with their scores as predicted by model.
  - double harassment
    
    The score for the category 'harassment'.
  - double harassmentThreatening
    
    The score for the category 'harassment/threatening'.
  - double hate
    
    The score for the category 'hate'.
  - double hateThreatening
    
    The score for the category 'hate/threatening'.
  - double illicit
    
    The score for the category 'illicit'.
  - double illicitViolent
    
    The score for the category 'illicit/violent'.
  - double selfHarm
    
    The score for the category 'self-harm'.
  - double selfHarmInstructions
    
    The score for the category 'self-harm/instructions'.
  - double selfHarmIntent
    
    The score for the category 'self-harm/intent'.
  - double sexual
    
    The score for the category 'sexual'.
  - double sexualMinors
    
    The score for the category 'sexual/minors'.
  - double violence
    
    The score for the category 'violence'.
  - double violenceGraphic
    
    The score for the category 'violence/graphic'.
- boolean flagged
  
  Whether any of the below categories are flagged.

Moderation Image URL Input

class ModerationImageUrlInput:

An object describing an image to classify.
- ImageUrl imageUrl
  
  Contains either an image URL or a data URL for a base64 encoded image.
  - String url
    
    Either a URL of the image or the base64 encoded image data.
- JsonValue; type "image_url"constant
  
  Always image_url.
  - IMAGE_URL("image_url")

Moderation Model

enum ModerationModel:
- OMNI_MODERATION_LATEST("omni-moderation-latest")
- OMNI_MODERATION_2024_09_26("omni-moderation-2024-09-26")
- TEXT_MODERATION_LATEST("text-moderation-latest")
- TEXT_MODERATION_STABLE("text-moderation-stable")

class ModerationMultiModalInput: A class that can be one of several variants.union

An object describing an image to classify.
- class ModerationImageUrlInput:
  
  An object describing an image to classify.
  - ImageUrl imageUrl
    
    Contains either an image URL or a data URL for a base64 encoded image.
    - String url
      
      Either a URL of the image or the base64 encoded image data.
  - JsonValue; type "image_url"constant
    
    Always image_url.
    - IMAGE_URL("image_url")
- class ModerationTextInput:
  
  An object describing text to classify.
  - String text
    
    A string of text to classify.
  - JsonValue; type "text"constant
    
    Always text.
    - TEXT("text")

Moderation Text Input

class ModerationTextInput:

An object describing text to classify.
- String text
  
  A string of text to classify.
- JsonValue; type "text"constant
  
  Always text.
  - TEXT("text")

java/resources/moderations/index.md +702 −0 created

1# Moderations

3## Create moderation

5`ModerationCreateResponse moderations().create(ModerationCreateParamsparams, RequestOptionsrequestOptions = RequestOptions.none())`

7**post** `/moderations`

9Classifies if text and/or image inputs are potentially harmful. Learn

10more in the [moderation guide](https://platform.openai.com/docs/guides/moderation).

12### Parameters

14- `ModerationCreateParams params`

16 - `Input input`

18 Input (or inputs) to classify. Can be a single string, an array of strings, or

19 an array of multi-modal input objects similar to other models.

21 - `String`

23 - `List<String>`

25 - `List<ModerationMultiModalInput>`

27 - `class ModerationImageUrlInput:`

29 An object describing an image to classify.

31 - `ImageUrl imageUrl`

33 Contains either an image URL or a data URL for a base64 encoded image.

35 - `String url`

37 Either a URL of the image or the base64 encoded image data.

39 - `JsonValue; type "image_url"constant`

41 Always `image_url`.

43 - `IMAGE_URL("image_url")`

45 - `class ModerationTextInput:`

47 An object describing text to classify.

49 - `String text`

51 A string of text to classify.

53 - `JsonValue; type "text"constant`

55 Always `text`.

57 - `TEXT("text")`

59 - `Optional<ModerationModel> model`

61 The content moderation model you would like to use. Learn more in

62 [the moderation guide](https://platform.openai.com/docs/guides/moderation), and learn about

63 available models [here](https://platform.openai.com/docs/models#moderation).

65### Returns

67- `class ModerationCreateResponse:`

69 Represents if a given text input is potentially harmful.

71 - `String id`

73 The unique identifier for the moderation request.

75 - `String model`

77 The model used to generate the moderation results.

79 - `List<Moderation> results`

81 A list of moderation objects.

83 - `Categories categories`

85 A list of the categories, and whether they are flagged or not.

87 - `boolean harassment`

89 Content that expresses, incites, or promotes harassing language towards any target.

91 - `boolean harassmentThreatening`

93 Harassment content that also includes violence or serious harm towards any target.

95 - `boolean hate`

97 Content that expresses, incites, or promotes hate based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste. Hateful content aimed at non-protected groups (e.g., chess players) is harassment.

99 - `boolean hateThreatening`

100

101 Hateful content that also includes violence or serious harm towards the targeted group based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste.

102

103 - `Optional<Boolean> illicit`

104

105 Content that includes instructions or advice that facilitate the planning or execution of wrongdoing, or that gives advice or instruction on how to commit illicit acts. For example, "how to shoplift" would fit this category.

106

107 - `Optional<Boolean> illicitViolent`

108

109 Content that includes instructions or advice that facilitate the planning or execution of wrongdoing that also includes violence, or that gives advice or instruction on the procurement of any weapon.

110

111 - `boolean selfHarm`

112

113 Content that promotes, encourages, or depicts acts of self-harm, such as suicide, cutting, and eating disorders.

114

115 - `boolean selfHarmInstructions`

116

117 Content that encourages performing acts of self-harm, such as suicide, cutting, and eating disorders, or that gives instructions or advice on how to commit such acts.

118

119 - `boolean selfHarmIntent`

120

121 Content where the speaker expresses that they are engaging or intend to engage in acts of self-harm, such as suicide, cutting, and eating disorders.

122

123 - `boolean sexual`

124

125 Content meant to arouse sexual excitement, such as the description of sexual activity, or that promotes sexual services (excluding sex education and wellness).

126

127 - `boolean sexualMinors`

128

129 Sexual content that includes an individual who is under 18 years old.

130

131 - `boolean violence`

132

133 Content that depicts death, violence, or physical injury.

134

135 - `boolean violenceGraphic`

136

137 Content that depicts death, violence, or physical injury in graphic detail.

138

139 - `CategoryAppliedInputTypes categoryAppliedInputTypes`

140

141 A list of the categories along with the input type(s) that the score applies to.

142

143 - `List<Harassment> harassment`

144

145 The applied input type(s) for the category 'harassment'.

146

147 - `TEXT("text")`

148

149 - `List<HarassmentThreatening> harassmentThreatening`

150

151 The applied input type(s) for the category 'harassment/threatening'.

152

153 - `TEXT("text")`

154

155 - `List<Hate> hate`

156

157 The applied input type(s) for the category 'hate'.

158

159 - `TEXT("text")`

160

161 - `List<HateThreatening> hateThreatening`

162

163 The applied input type(s) for the category 'hate/threatening'.

164

165 - `TEXT("text")`

166

167 - `List<Illicit> illicit`

168

169 The applied input type(s) for the category 'illicit'.

170

171 - `TEXT("text")`

172

173 - `List<IllicitViolent> illicitViolent`

174

175 The applied input type(s) for the category 'illicit/violent'.

176

177 - `TEXT("text")`

178

179 - `List<SelfHarm> selfHarm`

180

181 The applied input type(s) for the category 'self-harm'.

182

183 - `TEXT("text")`

184

185 - `IMAGE("image")`

186

187 - `List<SelfHarmInstruction> selfHarmInstructions`

188

189 The applied input type(s) for the category 'self-harm/instructions'.

190

191 - `TEXT("text")`

192

193 - `IMAGE("image")`

194

195 - `List<SelfHarmIntent> selfHarmIntent`

196

197 The applied input type(s) for the category 'self-harm/intent'.

198

199 - `TEXT("text")`

200

201 - `IMAGE("image")`

202

203 - `List<Sexual> sexual`

204

205 The applied input type(s) for the category 'sexual'.

206

207 - `TEXT("text")`

208

209 - `IMAGE("image")`

210

211 - `List<SexualMinor> sexualMinors`

212

213 The applied input type(s) for the category 'sexual/minors'.

214

215 - `TEXT("text")`

216

217 - `List<Violence> violence`

218

219 The applied input type(s) for the category 'violence'.

220

221 - `TEXT("text")`

222

223 - `IMAGE("image")`

224

225 - `List<ViolenceGraphic> violenceGraphic`

226

227 The applied input type(s) for the category 'violence/graphic'.

228

229 - `TEXT("text")`

230

231 - `IMAGE("image")`

232

233 - `CategoryScores categoryScores`

234

235 A list of the categories along with their scores as predicted by model.

236

237 - `double harassment`

238

239 The score for the category 'harassment'.

240

241 - `double harassmentThreatening`

242

243 The score for the category 'harassment/threatening'.

244

245 - `double hate`

246

247 The score for the category 'hate'.

248

249 - `double hateThreatening`

250

251 The score for the category 'hate/threatening'.

252

253 - `double illicit`

254

255 The score for the category 'illicit'.

256

257 - `double illicitViolent`

258

259 The score for the category 'illicit/violent'.

260

261 - `double selfHarm`

262

263 The score for the category 'self-harm'.

264

265 - `double selfHarmInstructions`

266

267 The score for the category 'self-harm/instructions'.

268

269 - `double selfHarmIntent`

270

271 The score for the category 'self-harm/intent'.

272

273 - `double sexual`

274

275 The score for the category 'sexual'.

276

277 - `double sexualMinors`

278

279 The score for the category 'sexual/minors'.

280

281 - `double violence`

282

283 The score for the category 'violence'.

284

285 - `double violenceGraphic`

286

287 The score for the category 'violence/graphic'.

288

289 - `boolean flagged`

290

291 Whether any of the below categories are flagged.

292

293### Example

294

295```java

296package com.openai.example;

297

298import com.openai.client.OpenAIClient;

299import com.openai.client.okhttp.OpenAIOkHttpClient;

300import com.openai.models.moderations.ModerationCreateParams;

301import com.openai.models.moderations.ModerationCreateResponse;

302

303public final class Main {

304 private Main() {}

305

306 public static void main(String[] args) {

307 OpenAIClient client = OpenAIOkHttpClient.fromEnv();

308

309 ModerationCreateParams params = ModerationCreateParams.builder()

310 .input("I want to kill them.")

311 .build();

312 ModerationCreateResponse moderation = client.moderations().create(params);

313 }

314}

315```

316

317#### Response

318

319```json

320{

321 "id": "id",

322 "model": "model",

323 "results": [

324 {

325 "categories": {

326 "harassment": true,

327 "harassment/threatening": true,

328 "hate": true,

329 "hate/threatening": true,

330 "illicit": true,

331 "illicit/violent": true,

332 "self-harm": true,

333 "self-harm/instructions": true,

334 "self-harm/intent": true,

335 "sexual": true,

336 "sexual/minors": true,

337 "violence": true,

338 "violence/graphic": true

339 },

340 "category_applied_input_types": {

341 "harassment": [

342 "text"

343 ],

344 "harassment/threatening": [

345 "text"

346 ],

347 "hate": [

348 "text"

349 ],

350 "hate/threatening": [

351 "text"

352 ],

353 "illicit": [

354 "text"

355 ],

356 "illicit/violent": [

357 "text"

358 ],

359 "self-harm": [

360 "text"

361 ],

362 "self-harm/instructions": [

363 "text"

364 ],

365 "self-harm/intent": [

366 "text"

367 ],

368 "sexual": [

369 "text"

370 ],

371 "sexual/minors": [

372 "text"

373 ],

374 "violence": [

375 "text"

376 ],

377 "violence/graphic": [

378 "text"

379 ]

380 },

381 "category_scores": {

382 "harassment": 0,

383 "harassment/threatening": 0,

384 "hate": 0,

385 "hate/threatening": 0,

386 "illicit": 0,

387 "illicit/violent": 0,

388 "self-harm": 0,

389 "self-harm/instructions": 0,

390 "self-harm/intent": 0,

391 "sexual": 0,

392 "sexual/minors": 0,

393 "violence": 0,

394 "violence/graphic": 0

395 },

396 "flagged": true

397 }

398 ]

399}

400```

401

402## Domain Types

403

404### Moderation

405

406- `class Moderation:`

407

408 - `Categories categories`

409

410 A list of the categories, and whether they are flagged or not.

411

412 - `boolean harassment`

413

414 Content that expresses, incites, or promotes harassing language towards any target.

415

416 - `boolean harassmentThreatening`

417

418 Harassment content that also includes violence or serious harm towards any target.

419

420 - `boolean hate`

421

422 Content that expresses, incites, or promotes hate based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste. Hateful content aimed at non-protected groups (e.g., chess players) is harassment.

423

424 - `boolean hateThreatening`

425

426 Hateful content that also includes violence or serious harm towards the targeted group based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste.

427

428 - `Optional<Boolean> illicit`

429

430 Content that includes instructions or advice that facilitate the planning or execution of wrongdoing, or that gives advice or instruction on how to commit illicit acts. For example, "how to shoplift" would fit this category.

431

432 - `Optional<Boolean> illicitViolent`

433

434 Content that includes instructions or advice that facilitate the planning or execution of wrongdoing that also includes violence, or that gives advice or instruction on the procurement of any weapon.

435

436 - `boolean selfHarm`

437

438 Content that promotes, encourages, or depicts acts of self-harm, such as suicide, cutting, and eating disorders.

439

440 - `boolean selfHarmInstructions`

441

442 Content that encourages performing acts of self-harm, such as suicide, cutting, and eating disorders, or that gives instructions or advice on how to commit such acts.

443

444 - `boolean selfHarmIntent`

445

446 Content where the speaker expresses that they are engaging or intend to engage in acts of self-harm, such as suicide, cutting, and eating disorders.

447

448 - `boolean sexual`

449

450 Content meant to arouse sexual excitement, such as the description of sexual activity, or that promotes sexual services (excluding sex education and wellness).

451

452 - `boolean sexualMinors`

453

454 Sexual content that includes an individual who is under 18 years old.

455

456 - `boolean violence`

457

458 Content that depicts death, violence, or physical injury.

459

460 - `boolean violenceGraphic`

461

462 Content that depicts death, violence, or physical injury in graphic detail.

463

464 - `CategoryAppliedInputTypes categoryAppliedInputTypes`

465

466 A list of the categories along with the input type(s) that the score applies to.

467

468 - `List<Harassment> harassment`

469

470 The applied input type(s) for the category 'harassment'.

471

472 - `TEXT("text")`

473

474 - `List<HarassmentThreatening> harassmentThreatening`

475

476 The applied input type(s) for the category 'harassment/threatening'.

477

478 - `TEXT("text")`

479

480 - `List<Hate> hate`

481

482 The applied input type(s) for the category 'hate'.

483

484 - `TEXT("text")`

485

486 - `List<HateThreatening> hateThreatening`

487

488 The applied input type(s) for the category 'hate/threatening'.

489

490 - `TEXT("text")`

491

492 - `List<Illicit> illicit`

493

494 The applied input type(s) for the category 'illicit'.

495

496 - `TEXT("text")`

497

498 - `List<IllicitViolent> illicitViolent`

499

500 The applied input type(s) for the category 'illicit/violent'.

501

502 - `TEXT("text")`

503

504 - `List<SelfHarm> selfHarm`

505

506 The applied input type(s) for the category 'self-harm'.

507

508 - `TEXT("text")`

509

510 - `IMAGE("image")`

511

512 - `List<SelfHarmInstruction> selfHarmInstructions`

513

514 The applied input type(s) for the category 'self-harm/instructions'.

515

516 - `TEXT("text")`

517

518 - `IMAGE("image")`

519

520 - `List<SelfHarmIntent> selfHarmIntent`

521

522 The applied input type(s) for the category 'self-harm/intent'.

523

524 - `TEXT("text")`

525

526 - `IMAGE("image")`

527

528 - `List<Sexual> sexual`

529

530 The applied input type(s) for the category 'sexual'.

531

532 - `TEXT("text")`

533

534 - `IMAGE("image")`

535

536 - `List<SexualMinor> sexualMinors`

537

538 The applied input type(s) for the category 'sexual/minors'.

539

540 - `TEXT("text")`

541

542 - `List<Violence> violence`

543

544 The applied input type(s) for the category 'violence'.

545

546 - `TEXT("text")`

547

548 - `IMAGE("image")`

549

550 - `List<ViolenceGraphic> violenceGraphic`

551

552 The applied input type(s) for the category 'violence/graphic'.

553

554 - `TEXT("text")`

555

556 - `IMAGE("image")`

557

558 - `CategoryScores categoryScores`

559

560 A list of the categories along with their scores as predicted by model.

561

562 - `double harassment`

563

564 The score for the category 'harassment'.

565

566 - `double harassmentThreatening`

567

568 The score for the category 'harassment/threatening'.

569

570 - `double hate`

571

572 The score for the category 'hate'.

573

574 - `double hateThreatening`

575

576 The score for the category 'hate/threatening'.

577

578 - `double illicit`

579

580 The score for the category 'illicit'.

581

582 - `double illicitViolent`

583

584 The score for the category 'illicit/violent'.

585

586 - `double selfHarm`

587

588 The score for the category 'self-harm'.

589

590 - `double selfHarmInstructions`

591

592 The score for the category 'self-harm/instructions'.

593

594 - `double selfHarmIntent`

595

596 The score for the category 'self-harm/intent'.

597

598 - `double sexual`

599

600 The score for the category 'sexual'.

601

602 - `double sexualMinors`

603

604 The score for the category 'sexual/minors'.

605

606 - `double violence`

607

608 The score for the category 'violence'.

609

610 - `double violenceGraphic`

611

612 The score for the category 'violence/graphic'.

613

614 - `boolean flagged`

615

616 Whether any of the below categories are flagged.

617

618### Moderation Image URL Input

619

620- `class ModerationImageUrlInput:`

621

622 An object describing an image to classify.

623

624 - `ImageUrl imageUrl`

625

626 Contains either an image URL or a data URL for a base64 encoded image.

627

628 - `String url`

629

630 Either a URL of the image or the base64 encoded image data.

631

632 - `JsonValue; type "image_url"constant`

633

634 Always `image_url`.

635

636 - `IMAGE_URL("image_url")`

637

638### Moderation Model

639

640- `enum ModerationModel:`

641

642 - `OMNI_MODERATION_LATEST("omni-moderation-latest")`

643

644 - `OMNI_MODERATION_2024_09_26("omni-moderation-2024-09-26")`

645

646 - `TEXT_MODERATION_LATEST("text-moderation-latest")`

647

648 - `TEXT_MODERATION_STABLE("text-moderation-stable")`

649

650### Moderation Multi Modal Input

651

652- `class ModerationMultiModalInput: A class that can be one of several variants.union`

653

654 An object describing an image to classify.

655

656 - `class ModerationImageUrlInput:`

657

658 An object describing an image to classify.

659

660 - `ImageUrl imageUrl`

661

662 Contains either an image URL or a data URL for a base64 encoded image.

663

664 - `String url`

665

666 Either a URL of the image or the base64 encoded image data.

667

668 - `JsonValue; type "image_url"constant`

669

670 Always `image_url`.

671

672 - `IMAGE_URL("image_url")`

673

674 - `class ModerationTextInput:`

675

676 An object describing text to classify.

677

678 - `String text`

679

680 A string of text to classify.

681

682 - `JsonValue; type "text"constant`

683

684 Always `text`.

685

686 - `TEXT("text")`

687

688### Moderation Text Input

689

690- `class ModerationTextInput:`

691

692 An object describing text to classify.

693

694 - `String text`

695

696 A string of text to classify.

697

698 - `JsonValue; type "text"constant`

699

700 Always `text`.

701

702 - `TEXT("text")`