Get Started
Examples
Concepts
Resources
Projects
Integrations
Precontext
copy markdown
Interfaze is a hybrid model architecture that combines multiple specialized smaller models adapted to work together as a single LLM.
These specialized layers generate additional high quality structured metadata like confidence scores, bounding boxes, etc. that can be accessed through the precontext field.
This allows you to build reliable applications with predictable and structured metadata you can use to validate the confidence and accuracy of the model's responses.
When is precontext available?
precontext is only available for prompts that requires the model to perform one of the following tasks:
- Vision (OCR, Object Detection)
- Audio (Speech-to-Text)
- Web (Web Search, Scraping)
- Translation
- Code (Code Sandboxing)
- Guardrails
Use can also use run tasks allows you to programmatically run parts of the model without activating the full model and getting the raw metadata as the full response making it faster and cheaper.
Response types
- Non-streaming responses: Returns
precontextby default - Streaming responses: Doesn't return
precontextby default (can be enabled by including thex-show-additional-infoheader)
Limits
Some SDKs don't provide a way to access the raw API response, so precontext is not available for those SDKs.
- LangChain SDK does not provide an official way to access raw API responses, so field natively is not available.
Precontext structure
Within the body of the chat completion response, precontext is an array of objects. Each item has a name stating the specific task that was performed and the result field containing the raw metadata for that task.
{
"object": {
"first_name": "IVÁN",
"last_name": "MUÑOZ ESTRADA",
"dob": "09/12/1987",
"driver_licence_number": "Y4067081"
},
"response": {
"id": "interfaze-1774997933544",
"modelId": "interfaze-beta",
"body": {
"id": "interfaze-1774997933544",
"object": "chat.completion",
"created": 1774997933,
"model": "interfaze-beta",
"usage": {
"prompt_tokens": 16771,
"completion_tokens": 2530,
"total_tokens": 19301
},
"precontext": [
{
"name": "ocr",
"result": {
"extracted_text": "California\nUSA\nDRIVER LICENSE\nDL Y4067081\nCLASS C\nEXP 09/12/2027\nEND NONE\nEN MUÑOZ ESTRADA\nFN IVÁN ICHET\n14223 BELGATE ST\nBALDWIN PARK CA 91706\nDOB 09/12/1987\nRSTRNONE 09121987\nIvan M.\nSEX M HAIR BLK EYES BLK\nHGT 5-02\" WGT 185lb ISS\nDD 08/13/202160648/AAFD/27 07/25/2022",
"sections": [
{
"text": "California\nUSA\nDRIVER LICENSE\nDL Y4067081\nCLASS C\nEXP 09/12/2027\nEND NONE\nEN MUÑOZ ESTRADA\nFN IVÁN ICHET\n14223 BELGATE ST\nBALDWIN PARK CA 91706\nDOB 09/12/1987\nRSTRNONE 09121987\nIvan M.\nSEX M HAIR BLK EYES BLK\nHGT 5-02\" WGT 185lb ISS\nDD 08/13/202160648/AAFD/27 07/25/2022",
"lines": [
{
"text": "California",
"bounds": {
"top_left": {
"x": 63,
"y": 89
},
"top_right": {
"x": 268,
"y": 89
},
"bottom_right": {
"x": 268,
"y": 129
},
"bottom_left": {
"x": 63,
"y": 129
},
"width": 205,
"height": 40
},
"average_confidence": 0.99,
"words": [
{
"text": "California",
"bounds": {
"top_left": {
"x": 64,
"y": 90
},
"top_right": {
"x": 267,
"y": 89
},
"bottom_right": {
"x": 267,
"y": 130
},
"bottom_left": {
"x": 63,
"y": 130
},
"width": 203.5,
"height": 40.5
},
"confidence": 0.99
}
]
},
{
"text": "USA",
"bounds": {
"top_left": {
"x": 270,
"y": 100
},
"top_right": {
"x": 296,
"y": 99
},
"bottom_right": {
"x": 296,
"y": 114
},
"bottom_left": {
"x": 269,
"y": 115
},
"width": 26.5,
"height": 15
},
"average_confidence": 0.96,
"words": [
{
"text": "USA",
"bounds": {
"top_left": {
"x": 269,
"y": 100
},
"top_right": {
"x": 294,
"y": 99
},
"bottom_right": {
"x": 295,
"y": 114
},
"bottom_left": {
"x": 269,
"y": 115
},
"width": 25.5,
"height": 15
},
"confidence": 0.96
}
]
},
{
"text": "DRIVER LICENSE",
"bounds": {
"top_left": {
"x": 334,
"y": 91
},
"top_right": {
"x": 497,
"y": 91
},
"bottom_right": {
"x": 497,
"y": 113
},
"bottom_left": {
"x": 334,
"y": 114
},
"width": 163,
"height": 22.5
},
"average_confidence": 0.99,
"words": [
{
"text": "DRIVER",
"bounds": {
"top_left": {
"x": 335,
"y": 91
},
"top_right": {
"x": 405,
"y": 92
},
"bottom_right": {
"x": 405,
"y": 114
},
"bottom_left": {
"x": 335,
"y": 115
},
"width": 70,
"height": 23
},
"confidence": 0.99
},
{
"text": "LICENSE",
"bounds": {
"top_left": {
"x": 412,
"y": 92
},
"top_right": {
"x": 494,
"y": 91
},
"bottom_right": {
"x": 494,
"y": 113
},
"bottom_left": {
"x": 412,
"y": 114
},
"width": 82,
"height": 22
},
"confidence": 0.99
}
]
},
{
"text": "DL Y4067081",
"bounds": {
"top_left": {
"x": 247,
"y": 153
},
"top_right": {
"x": 394,
"y": 153
},
"bottom_right": {
"x": 394,
"y": 178
},
"bottom_left": {
"x": 247,
"y": 179
},
"width": 147,
"height": 25.5
},
"average_confidence": 0.93,
"words": [
{
"text": "DL",
"bounds": {
"top_left": {
"x": 247,
"y": 154
},
"top_right": {
"x": 267,
"y": 154
},
"bottom_right": {
"x": 267,
"y": 179
},
"bottom_left": {
"x": 247,
"y": 179
},
"width": 20,
"height": 25
},
"confidence": 0.92
},
{
"text": "Y4067081",
"bounds": {
"top_left": {
"x": 272,
"y": 154
},
"top_right": {
"x": 393,
"y": 154
},
"bottom_right": {
"x": 393,
"y": 179
},
"bottom_left": {
"x": 272,
"y": 179
},
"width": 121,
"height": 25
},
"confidence": 0.95
}
]
},
{
"text": "CLASS C",
"bounds": {
"top_left": {
"x": 453,
"y": 153
},
"top_right": {
"x": 521,
"y": 153
},
"bottom_right": {
"x": 521,
"y": 168
},
"bottom_left": {
"x": 453,
"y": 169
},
"width": 68,
"height": 15.5
},
"average_confidence": 0.99,
"words": [
{
"text": "CLASS",
"bounds": {
"top_left": {
"x": 454,
"y": 153
},
"top_right": {
"x": 503,
"y": 153
},
"bottom_right": {
"x": 503,
"y": 169
},
"bottom_left": {
"x": 455,
"y": 169
},
"width": 48.5,
"height": 16
},
"confidence": 0.99
},
{
"text": "C",
"bounds": {
"top_left": {
"x": 510,
"y": 154
},
"top_right": {
"x": 518,
"y": 154
},
"bottom_right": {
"x": 518,
"y": 168
},
"bottom_left": {
"x": 510,
"y": 169
},
"width": 8,
"height": 14.5
},
"confidence": 0.99
}
]
},
{
"text": "EXP 09/12/2027",
"bounds": {
"top_left": {
"x": 244,
"y": 184
},
"top_right": {
"x": 397,
"y": 184
},
"bottom_right": {
"x": 397,
"y": 207
},
"bottom_left": {
"x": 244,
"y": 208
},
"width": 153,
"height": 23.5
},
"average_confidence": 0.96,
"words": [
{
"text": "EXP",
"bounds": {
"top_left": {
"x": 249,
"y": 185
},
"top_right": {
"x": 278,
"y": 185
},
"bottom_right": {
"x": 277,
"y": 209
},
"bottom_left": {
"x": 248,
"y": 208
},
"width": 29,
"height": 23.5
},
"confidence": 0.97
},
{
"text": "09/12/2027",
"bounds": {
"top_left": {
"x": 283,
"y": 185
},
"top_right": {
"x": 397,
"y": 184
},
"bottom_right": {
"x": 397,
"y": 208
},
"bottom_left": {
"x": 282,
"y": 209
},
"width": 114.5,
"height": 24
},
"confidence": 0.94
}
]
},
{
"text": "END NONE",
"bounds": {
"top_left": {
"x": 454,
"y": 191
},
"top_right": {
"x": 536,
"y": 191
},
"bottom_right": {
"x": 536,
"y": 206
},
"bottom_left": {
"x": 454,
"y": 206
},
"width": 82,
"height": 15
},
"average_confidence": 0.99,
"words": [
{
"text": "END",
"bounds": {
"top_left": {
"x": 456,
"y": 192
},
"top_right": {
"x": 484,
"y": 192
},
"bottom_right": {
"x": 484,
"y": 206
},
"bottom_left": {
"x": 456,
"y": 206
},
"width": 28,
"height": 14
},
"confidence": 0.99
},
{
"text": "NONE",
"bounds": {
"top_left": {
"x": 493,
"y": 192
},
"top_right": {
"x": 533,
"y": 192
},
"bottom_right": {
"x": 533,
"y": 207
},
"bottom_left": {
"x": 492,
"y": 207
},
"width": 40.5,
"height": 15
},
"confidence": 0.99
}
]
},
{
"text": "EN MUÑOZ ESTRADA",
"bounds": {
"top_left": {
"x": 248,
"y": 213
},
"top_right": {
"x": 449,
"y": 211
},
"bottom_right": {
"x": 449,
"y": 232
},
"bottom_left": {
"x": 248,
"y": 234
},
"width": 201,
"height": 21
},
"average_confidence": 0.73,
"words": [
{
"text": "EN",
"bounds": {
"top_left": {
"x": 249,
"y": 214
},
"top_right": {
"x": 269,
"y": 213
},
"bottom_right": {
"x": 269,
"y": 234
},
"bottom_left": {
"x": 249,
"y": 234
},
"width": 20,
"height": 20.5
},
"confidence": 0.31
},
{
"text": "MUÑOZ",
"bounds": {
"top_left": {
"x": 274,
"y": 213
},
"top_right": {
"x": 347,
"y": 212
},
"bottom_right": {
"x": 347,
"y": 234
},
"bottom_left": {
"x": 274,
"y": 234
},
"width": 73,
"height": 21.5
},
"confidence": 0.95
},
{
"text": "ESTRADA",
"bounds": {
"top_left": {
"x": 353,
"y": 212
},
"top_right": {
"x": 444,
"y": 212
},
"bottom_right": {
"x": 444,
"y": 233
},
"bottom_left": {
"x": 353,
"y": 234
},
"width": 91,
"height": 21.5
},
"confidence": 0.92
}
]
},
{
"text": "FN IVÁN ICHET",
"bounds": {
"top_left": {
"x": 250,
"y": 233
},
"top_right": {
"x": 389,
"y": 233
},
"bottom_right": {
"x": 389,
"y": 253
},
"bottom_left": {
"x": 250,
"y": 254
},
"width": 139,
"height": 20.5
},
"average_confidence": 0.85,
"words": [
{
"text": "FN",
"bounds": {
"top_left": {
"x": 250,
"y": 234
},
"top_right": {
"x": 270,
"y": 234
},
"bottom_right": {
"x": 270,
"y": 254
},
"bottom_left": {
"x": 250,
"y": 255
},
"width": 20,
"height": 20.5
},
"confidence": 0.86
},
{
"text": "IVÁN",
"bounds": {
"top_left": {
"x": 274,
"y": 234
},
"top_right": {
"x": 318,
"y": 234
},
"bottom_right": {
"x": 317,
"y": 254
},
"bottom_left": {
"x": 274,
"y": 254
},
"width": 43.5,
"height": 20
},
"confidence": 0.7
},
{
"text": "ICHET",
"bounds": {
"top_left": {
"x": 327,
"y": 234
},
"top_right": {
"x": 387,
"y": 233
},
"bottom_right": {
"x": 386,
"y": 254
},
"bottom_left": {
"x": 327,
"y": 254
},
"width": 59.5,
"height": 20.5
},
"confidence": 0.99
}
]
},
{
"text": "14223 BELGATE ST",
"bounds": {
"top_left": {
"x": 250,
"y": 255
},
"top_right": {
"x": 393,
"y": 253
},
"bottom_right": {
"x": 393,
"y": 271
},
"bottom_left": {
"x": 251,
"y": 272
},
"width": 142.5,
"height": 17.5
},
"average_confidence": 0.99,
"words": [
{
"text": "14223",
"bounds": {
"top_left": {
"x": 252,
"y": 256
},
"top_right": {
"x": 293,
"y": 256
},
"bottom_right": {
"x": 293,
"y": 272
},
"bottom_left": {
"x": 252,
"y": 272
},
"width": 41,
"height": 16
},
"confidence": 1
},
{
"text": "BELGATE",
"bounds": {
"top_left": {
"x": 296,
"y": 256
},
"top_right": {
"x": 366,
"y": 255
},
"bottom_right": {
"x": 366,
"y": 271
},
"bottom_left": {
"x": 297,
"y": 272
},
"width": 69.5,
"height": 16
},
"confidence": 0.99
},
{
"text": "ST",
"bounds": {
"top_left": {
"x": 373,
"y": 254
},
"top_right": {
"x": 391,
"y": 254
},
"bottom_right": {
"x": 391,
"y": 271
},
"bottom_left": {
"x": 372,
"y": 271
},
"width": 18.5,
"height": 17
},
"confidence": 0.99
}
]
},
{
"text": "BALDWIN PARK CA 91706",
"bounds": {
"top_left": {
"x": 250,
"y": 270
},
"top_right": {
"x": 440,
"y": 269
},
"bottom_right": {
"x": 440,
"y": 286
},
"bottom_left": {
"x": 250,
"y": 287
},
"width": 190,
"height": 17
},
"average_confidence": 0.99,
"words": [
{
"text": "BALDWIN",
"bounds": {
"top_left": {
"x": 251,
"y": 271
},
"top_right": {
"x": 320,
"y": 270
},
"bottom_right": {
"x": 319,
"y": 287
},
"bottom_left": {
"x": 251,
"y": 287
},
"width": 68.5,
"height": 16.5
},
"confidence": 0.99
},
{
"text": "PARK",
"bounds": {
"top_left": {
"x": 327,
"y": 270
},
"top_right": {
"x": 365,
"y": 270
},
"bottom_right": {
"x": 365,
"y": 287
},
"bottom_left": {
"x": 326,
"y": 287
},
"width": 38.5,
"height": 17
},
"confidence": 0.99
},
{
"text": "CA",
"bounds": {
"top_left": {
"x": 372,
"y": 270
},
"top_right": {
"x": 392,
"y": 270
},
"bottom_right": {
"x": 392,
"y": 287
},
"bottom_left": {
"x": 372,
"y": 287
},
"width": 20,
"height": 17
},
"confidence": 1
},
{
"text": "91706",
"bounds": {
"top_left": {
"x": 397,
"y": 270
},
"top_right": {
"x": 439,
"y": 269
},
"bottom_right": {
"x": 438,
"y": 287
},
"bottom_left": {
"x": 396,
"y": 287
},
"width": 42,
"height": 17.5
},
"confidence": 0.99
}
]
},
{
"text": "DOB 09/12/1987",
"bounds": {
"top_left": {
"x": 251,
"y": 291
},
"top_right": {
"x": 404,
"y": 289
},
"bottom_right": {
"x": 404,
"y": 311
},
"bottom_left": {
"x": 251,
"y": 313
},
"width": 153,
"height": 22
},
"average_confidence": 0.99,
"words": [
{
"text": "DOB",
"bounds": {
"top_left": {
"x": 252,
"y": 293
},
"top_right": {
"x": 285,
"y": 292
},
"bottom_right": {
"x": 285,
"y": 312
},
"bottom_left": {
"x": 252,
"y": 313
},
"width": 33,
"height": 20
},
"confidence": 1
},
{
"text": "09/12/1987",
"bounds": {
"top_left": {
"x": 291,
"y": 291
},
"top_right": {
"x": 404,
"y": 290
},
"bottom_right": {
"x": 404,
"y": 312
},
"bottom_left": {
"x": 291,
"y": 312
},
"width": 113,
"height": 21.5
},
"confidence": 0.99
}
]
},
{
"text": "RSTRNONE 09121987",
"bounds": {
"top_left": {
"x": 251,
"y": 314
},
"top_right": {
"x": 343,
"y": 314
},
"bottom_right": {
"x": 343,
"y": 330
},
"bottom_left": {
"x": 251,
"y": 331
},
"width": 92,
"height": 16.5
},
"average_confidence": 0.96,
"words": [
{
"text": "RSTRNONE",
"bounds": {
"top_left": {
"x": 252,
"y": 315
},
"top_right": {
"x": 341,
"y": 314
},
"bottom_right": {
"x": 341,
"y": 331
},
"bottom_left": {
"x": 251,
"y": 331
},
"width": 89.5,
"height": 16.5
},
"confidence": 0.96
}
]
},
{
"text": "Ivan M.",
"bounds": {
"top_left": {
"x": 76,
"y": 374
},
"top_right": {
"x": 239,
"y": 376
},
"bottom_right": {
"x": 239,
"y": 420
},
"bottom_left": {
"x": 75,
"y": 419
},
"width": 163.5,
"height": 44.5
},
"average_confidence": 0.9,
"words": [
{
"text": "Ivan",
"bounds": {
"top_left": {
"x": 85,
"y": 375
},
"top_right": {
"x": 172,
"y": 376
},
"bottom_right": {
"x": 171,
"y": 419
},
"bottom_left": {
"x": 83,
"y": 420
},
"width": 87.5,
"height": 44
},
"confidence": 0.94
},
{
"text": "M.",
"bounds": {
"top_left": {
"x": 199,
"y": 377
},
"top_right": {
"x": 237,
"y": 379
},
"bottom_right": {
"x": 237,
"y": 420
},
"bottom_left": {
"x": 198,
"y": 420
},
"width": 38.5,
"height": 42
},
"confidence": 0.87
}
]
},
{
"text": "SEX M HAIR BLK EYES BLK",
"bounds": {
"top_left": {
"x": 315,
"y": 369
},
"top_right": {
"x": 370,
"y": 369
},
"bottom_right": {
"x": 370,
"y": 384
},
"bottom_left": {
"x": 315,
"y": 384
},
"width": 55,
"height": 15
},
"average_confidence": 0.99,
"words": [
{
"text": "SEX",
"bounds": {
"top_left": {
"x": 317,
"y": 370
},
"top_right": {
"x": 345,
"y": 370
},
"bottom_right": {
"x": 345,
"y": 385
},
"bottom_left": {
"x": 317,
"y": 385
},
"width": 28,
"height": 15
},
"confidence": 0.98
},
{
"text": "M",
"bounds": {
"top_left": {
"x": 356,
"y": 370
},
"top_right": {
"x": 364,
"y": 370
},
"bottom_right": {
"x": 364,
"y": 385
},
"bottom_left": {
"x": 356,
"y": 385
},
"width": 8,
"height": 15
},
"confidence": 1
}
]
},
{
"text": "HGT 5-02\" WGT 185lb ISS",
"bounds": {
"top_left": {
"x": 316,
"y": 387
},
"top_right": {
"x": 400,
"y": 387
},
"bottom_right": {
"x": 400,
"y": 401
},
"bottom_left": {
"x": 316,
"y": 402
},
"width": 84,
"height": 14.5
},
"average_confidence": 0.82,
"words": [
{
"text": "HGT",
"bounds": {
"top_left": {
"x": 317,
"y": 388
},
"top_right": {
"x": 347,
"y": 387
},
"bottom_right": {
"x": 346,
"y": 402
},
"bottom_left": {
"x": 317,
"y": 402
},
"width": 29.5,
"height": 14.5
},
"confidence": 0.99
},
{
"text": "5-02\"",
"bounds": {
"top_left": {
"x": 359,
"y": 387
},
"top_right": {
"x": 399,
"y": 387
},
"bottom_right": {
"x": 399,
"y": 402
},
"bottom_left": {
"x": 359,
"y": 401
},
"width": 40,
"height": 14.5
},
"confidence": 0.65
}
]
},
{
"text": "DD 08/13/202160648/AAFD/27 07/25/2022",
"bounds": {
"top_left": {
"x": 316,
"y": 408
},
"top_right": {
"x": 499,
"y": 406
},
"bottom_right": {
"x": 499,
"y": 421
},
"bottom_left": {
"x": 316,
"y": 423
},
"width": 183,
"height": 15
},
"average_confidence": 0.95,
"words": [
{
"text": "DD",
"bounds": {
"top_left": {
"x": 318,
"y": 408
},
"top_right": {
"x": 335,
"y": 408
},
"bottom_right": {
"x": 336,
"y": 423
},
"bottom_left": {
"x": 318,
"y": 423
},
"width": 17.5,
"height": 15
},
"confidence": 0.96
},
{
"text": "08/13/202160648/AAFD/27",
"bounds": {
"top_left": {
"x": 342,
"y": 408
},
"top_right": {
"x": 499,
"y": 407
},
"bottom_right": {
"x": 499,
"y": 420
},
"bottom_left": {
"x": 343,
"y": 423
},
"width": 156.5,
"height": 14
},
"confidence": 0.94
}
]
}
]
}
],
"width": 698,
"height": 525
}
}
]
}
},
"finishReason": "stop",
"usage": {
"inputTokens": 16771,
"outputTokens": 2530,
"totalTokens": 19301
}
}name: The name of the specialized task ran (e.g.,ocr,object_detection,speech_to_text)precontextcan contain multiple items of the same task name if the task was performed multiple timesresult: The raw metadata for that task- Each
resultis task specific and contains the raw metadata for that task. - Every task has a different result structure but the structure is fixed and predictable for each task type.
STT & Translation Precontext Example
Note: LangChain SDK does not provide an official way to access raw API responses, so we cannot include an example for that SDK at this time.
OpenAI SDK
Vercel AI SDK
LangChain SDK
typescript
python
import { z } from "zod";
import { zodResponseFormat } from "openai/helpers/zod";
const STTSchema = z.object({
translated_text: z.string().describe("translated text"),
original_language_code: z.string(),
translated_language_code: z.string(),
});
const response = await interfaze.chat.completions.create({
model: "interfaze-beta",
messages: [
{
role: "user",
content: "Transcribe the audio file and translate it to chinese https://r2public.jigsawstack.com/interfaze/examples/stt_medical_short.mp4",
},
],
response_format: zodResponseFormat(STTSchema, "stt_schema"),
});
console.log(response.choices[0].message.content);
//@ts-expect-error precontext is not typed
const precontext = response.precontext;
console.log("STT Results:", precontext?.[0]?.result);In this example, we're transcribing an audio file and translating the text to Chinese. You will notice two items in the precontext array, one for the STT task and one for the translation task along side your final response content.
{
"object": {
"translated_text": "我刚开始服用一轮阿莫西林，想问一下把它和我目前的螺内酯处方一起服用是否安全。",
"original_language_code": "en",
"translated_language_code": "zh"
},
"response": {
"id": "interfaze-1775088833045",
"modelId": "interfaze-beta",
"body": {
"id": "interfaze-1775088833045",
"object": "chat.completion",
"model": "interfaze-beta",
"usage": {
"prompt_tokens": 2758,
"completion_tokens": 800,
"total_tokens": 3558
},
"precontext": [
{
"name": "stt",
"result": {
"text": "I just started a round of amoxicillin and I wanted to ask if it was safe to take that with my current spironolactone prescription.",
"chunks": [
{
"timestamp": [0.28, 4],
"text": "I just started a round of amoxicillin and I wanted to ask"
},
{
"timestamp": [4, 7.72],
"text": "if it was safe to take that with my current spironolactone prescription."
}
]
}
},
{
"name": "translate",
"result": {
"translated_text": "我刚开始服用一轮阿莫西林，想问一下把它和我目前的螺内酯处方一起服用是否安全。",
"source_language": "auto-detected",
"target_language": "zh",
"batch_size": 1
}
}
]
}
},
"finishReason": "stop",
"usage": {
"inputTokens": 2758,
"outputTokens": 800,
"totalTokens": 3558
}
}Enabling Precontext in streaming responses
By default, streaming responses don't return precontext since the client will not be able to process the data separately from the main response.
However, you can enable it by including the x-show-additional-info header in your API requests.
This will stream the precontext data as a single chunk before the main response begins streaming.
headers: {
"x-show-additional-info": "true"
}In streaming mode, the precontext data is sent as a single chunk before the main response begins streaming. This chunk follows a structured format that encapsulates all pre-processing information upfront.
<precontext>
{
"name": "process_name",
"result": { /* detailed process output */ }
}
</precontext>Getting Precontext as the main response
You can use run tasks to run a specific task without having to run the full model making it faster and cheaper. However it is limited to the raw output provided in precontext with a fixed predefined structured output.
Learn more about running a task.