-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy path_client.py
More file actions
1109 lines (877 loc) · 40.3 KB
/
_client.py
File metadata and controls
1109 lines (877 loc) · 40.3 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
from __future__ import annotations
import os
from typing import (
TYPE_CHECKING,
Any,
Union,
Literal,
Mapping,
TypeVar,
Callable,
Iterator,
Optional,
AsyncIterator,
overload,
)
from typing_extensions import Self, Unpack, ParamSpec, override
import httpx
from replicate.lib.cog import _get_api_token_from_environment
from replicate.lib._files import FileEncodingStrategy
from replicate.lib._predictions_run import Model, Version, ModelVersionIdentifier
from replicate.types.prediction_create_params import PredictionCreateParamsWithoutVersion
from . import _exceptions
from ._qs import Querystring
from .types import client_search_params
from ._types import (
NOT_GIVEN,
Body,
Omit,
Query,
Headers,
Timeout,
NotGiven,
Transport,
ProxiesTypes,
RequestOptions,
)
from ._utils import (
is_given,
maybe_transform,
get_async_library,
async_maybe_transform,
)
from ._compat import cached_property
from ._version import __version__
from ._response import (
to_raw_response_wrapper,
to_streamed_response_wrapper,
async_to_raw_response_wrapper,
async_to_streamed_response_wrapper,
)
from ._streaming import Stream as Stream, AsyncStream as AsyncStream
from ._exceptions import APIStatusError, ReplicateError
from ._base_client import (
DEFAULT_MAX_RETRIES,
SyncAPIClient,
AsyncAPIClient,
make_request_options,
)
from .types.search_response import SearchResponse
if TYPE_CHECKING:
from .resources import files, models, account, hardware, webhooks, trainings, collections, deployments, predictions
from .resources.files import FilesResource, AsyncFilesResource
from .resources.account import AccountResource, AsyncAccountResource
from .resources.hardware import HardwareResource, AsyncHardwareResource
from .resources.trainings import TrainingsResource, AsyncTrainingsResource
from .resources.collections import CollectionsResource, AsyncCollectionsResource
from .resources.predictions import PredictionsResource, AsyncPredictionsResource
from .resources.models.models import ModelsResource, AsyncModelsResource
from .resources.webhooks.webhooks import WebhooksResource, AsyncWebhooksResource
from .resources.deployments.deployments import DeploymentsResource, AsyncDeploymentsResource
if TYPE_CHECKING:
from .lib._predictions_use import Function, FunctionRef, AsyncFunction
Input = ParamSpec("Input")
Output = TypeVar("Output")
__all__ = [
"Timeout",
"Transport",
"ProxiesTypes",
"RequestOptions",
"Replicate",
"AsyncReplicate",
"Client",
"AsyncClient",
]
class Replicate(SyncAPIClient):
# client options
bearer_token: str
def __init__(
self,
*,
bearer_token: str | None = None,
base_url: str | httpx.URL | None = None,
timeout: Union[float, Timeout, None, NotGiven] = NOT_GIVEN,
max_retries: int = DEFAULT_MAX_RETRIES,
default_headers: Mapping[str, str] | None = None,
default_query: Mapping[str, object] | None = None,
# Configure a custom httpx client.
# We provide a `DefaultHttpxClient` class that you can pass to retain the default values we use for `limits`, `timeout` & `follow_redirects`.
# See the [httpx documentation](https://www.python-httpx.org/api/#client) for more details.
http_client: httpx.Client | None = None,
# Enable or disable schema validation for data returned by the API.
# When enabled an error APIResponseValidationError is raised
# if the API responds with invalid data for the expected schema.
#
# This parameter may be removed or changed in the future.
# If you rely on this feature, please open a GitHub issue
# outlining your use-case to help us decide if it should be
# part of our public interface in the future.
_strict_response_validation: bool = False,
) -> None:
"""Construct a new synchronous Replicate client instance.
This automatically infers the `bearer_token` argument from the `REPLICATE_API_TOKEN` environment variable if it is not provided.
"""
if bearer_token is None:
bearer_token = _get_api_token_from_environment()
if bearer_token is None:
raise ReplicateError(
"The bearer_token client option must be set either by passing bearer_token to the client or by setting the REPLICATE_API_TOKEN environment variable"
)
self.bearer_token = bearer_token
if base_url is None:
base_url = os.environ.get("REPLICATE_BASE_URL")
if base_url is None:
base_url = f"https://api.replicate.com/v1"
super().__init__(
version=__version__,
base_url=base_url,
max_retries=max_retries,
timeout=timeout,
http_client=http_client,
custom_headers=default_headers,
custom_query=default_query,
_strict_response_validation=_strict_response_validation,
)
@cached_property
def collections(self) -> CollectionsResource:
from .resources.collections import CollectionsResource
return CollectionsResource(self)
@cached_property
def deployments(self) -> DeploymentsResource:
from .resources.deployments import DeploymentsResource
return DeploymentsResource(self)
@cached_property
def hardware(self) -> HardwareResource:
from .resources.hardware import HardwareResource
return HardwareResource(self)
@cached_property
def account(self) -> AccountResource:
from .resources.account import AccountResource
return AccountResource(self)
@cached_property
def models(self) -> ModelsResource:
from .resources.models import ModelsResource
return ModelsResource(self)
@cached_property
def predictions(self) -> PredictionsResource:
from .resources.predictions import PredictionsResource
return PredictionsResource(self)
@cached_property
def trainings(self) -> TrainingsResource:
from .resources.trainings import TrainingsResource
return TrainingsResource(self)
@cached_property
def webhooks(self) -> WebhooksResource:
from .resources.webhooks import WebhooksResource
return WebhooksResource(self)
@cached_property
def files(self) -> FilesResource:
from .resources.files import FilesResource
return FilesResource(self)
@cached_property
def with_raw_response(self) -> ReplicateWithRawResponse:
return ReplicateWithRawResponse(self)
@cached_property
def with_streaming_response(self) -> ReplicateWithStreamedResponse:
return ReplicateWithStreamedResponse(self)
@cached_property
def poll_interval(self) -> float:
return float(os.environ.get("REPLICATE_POLL_INTERVAL", "0.5"))
@property
@override
def qs(self) -> Querystring:
return Querystring(array_format="comma")
@property
@override
def auth_headers(self) -> dict[str, str]:
bearer_token = self.bearer_token
return {"Authorization": f"Bearer {bearer_token}"}
@property
@override
def default_headers(self) -> dict[str, str | Omit]:
return {
**super().default_headers,
"X-Stainless-Async": "false",
**self._custom_headers,
}
def run(
self,
ref: Union[Model, Version, ModelVersionIdentifier, str],
*,
file_encoding_strategy: Optional["FileEncodingStrategy"] = None,
use_file_output: bool = True,
wait: Union[int, bool, NotGiven] = NOT_GIVEN,
**params: Unpack[PredictionCreateParamsWithoutVersion],
) -> Any:
"""
Run a model prediction.
Args:
ref: Reference to the model or version to run. Can be:
- A string containing a version ID (e.g. "5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa")
- A string with owner/name format (e.g. "replicate/hello-world")
- A string with owner/name:version format (e.g. "replicate/hello-world:5c7d5dc6...")
- A Model instance with owner and name attributes
- A Version instance with id attribute
- A ModelVersionIdentifier dictionary with owner, name, and/or version keys
file_encoding_strategy: Strategy for encoding file inputs, options are "base64" or "url"
use_file_output: If True (default), convert output URLs to FileOutput objects
wait: If True (default), wait for the prediction to complete. If False, return immediately.
If an integer, wait up to that many seconds.
**params: Additional parameters to pass to the prediction creation endpoint including
the required "input" dictionary with model-specific parameters
Returns:
The prediction output, which could be a basic type (str, int, etc.), a FileOutput object,
a list of FileOutput objects, or a dictionary of FileOutput objects, depending on what
the model returns.
Raises:
ModelError: If the model run fails
ValueError: If the reference format is invalid
TypeError: If both wait and prefer parameters are provided
"""
from .lib._predictions_run import run
return run(
self,
ref,
wait=wait,
use_file_output=use_file_output,
file_encoding_strategy=file_encoding_strategy,
**params,
)
@overload
def use(
self,
ref: Union[str, "FunctionRef[Input, Output]"],
*,
hint: Optional[Callable["Input", "Output"]] = None,
streaming: Literal[False] = False,
) -> "Function[Input, Output]": ...
@overload
def use(
self,
ref: Union[str, "FunctionRef[Input, Output]"],
*,
hint: Optional[Callable["Input", "Output"]] = None,
streaming: Literal[True],
) -> "Function[Input, Iterator[Output]]": ...
def use(
self,
ref: Union[str, "FunctionRef[Input, Output]"],
*,
hint: Optional[Callable["Input", "Output"]] = None,
streaming: bool = False,
) -> Union["Function[Input, Output]", "Function[Input, Iterator[Output]]"]:
"""
Use a Replicate model as a function.
Example:
flux_dev = replicate.use("black-forest-labs/flux-dev")
output = flux_dev(prompt="make me a sandwich")
"""
from .lib._predictions_use import use as _use
# TODO: Fix mypy overload matching for streaming parameter
return _use(self, ref, hint=hint, streaming=streaming) # type: ignore[call-overload, no-any-return]
def copy(
self,
*,
bearer_token: str | None = None,
base_url: str | httpx.URL | None = None,
timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
http_client: httpx.Client | None = None,
max_retries: int | NotGiven = NOT_GIVEN,
default_headers: Mapping[str, str] | None = None,
set_default_headers: Mapping[str, str] | None = None,
default_query: Mapping[str, object] | None = None,
set_default_query: Mapping[str, object] | None = None,
_extra_kwargs: Mapping[str, Any] = {},
) -> Self:
"""
Create a new client instance re-using the same options given to the current client with optional overriding.
"""
if default_headers is not None and set_default_headers is not None:
raise ValueError("The `default_headers` and `set_default_headers` arguments are mutually exclusive")
if default_query is not None and set_default_query is not None:
raise ValueError("The `default_query` and `set_default_query` arguments are mutually exclusive")
headers = self._custom_headers
if default_headers is not None:
headers = {**headers, **default_headers}
elif set_default_headers is not None:
headers = set_default_headers
params = self._custom_query
if default_query is not None:
params = {**params, **default_query}
elif set_default_query is not None:
params = set_default_query
http_client = http_client or self._client
return self.__class__(
bearer_token=bearer_token or self.bearer_token,
base_url=base_url or self.base_url,
timeout=self.timeout if isinstance(timeout, NotGiven) else timeout,
http_client=http_client,
max_retries=max_retries if is_given(max_retries) else self.max_retries,
default_headers=headers,
default_query=params,
**_extra_kwargs,
)
# Alias for `copy` for nicer inline usage, e.g.
# client.with_options(timeout=10).foo.create(...)
with_options = copy
def search(
self,
*,
query: str,
limit: int | NotGiven = NOT_GIVEN,
# Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
# The extra values given here take precedence over values defined on the client or passed to this method.
extra_headers: Headers | None = None,
extra_query: Query | None = None,
extra_body: Body | None = None,
timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
) -> SearchResponse:
"""
Search for public models, collections, and docs using a text query.
For models, the response includes all model data, plus a new `metadata` object
with the following fields:
- `generated_description`: A longer and more detailed AI-generated description
of the model
- `tags`: An array of tags for the model
- `score`: A score for the model's relevance to the search query
Example cURL request:
```console
curl -s \\
-H "Authorization: Bearer $REPLICATE_API_TOKEN" \\
"https://api.replicate.com/v1/search?query=nano+banana"
```
Note: This search API is currently in beta and may change in future versions.
Args:
query: The search query string
limit: Maximum number of model results to return (1-50, defaults to 20)
extra_headers: Send extra headers
extra_query: Add additional query parameters to the request
extra_body: Add additional JSON properties to the request
timeout: Override the client-level default timeout for this request, in seconds
"""
return self.get(
"/search",
options=make_request_options(
extra_headers=extra_headers,
extra_query=extra_query,
extra_body=extra_body,
timeout=timeout,
query=maybe_transform(
{
"query": query,
"limit": limit,
},
client_search_params.ClientSearchParams,
),
),
cast_to=SearchResponse,
)
@override
def _make_status_error(
self,
err_msg: str,
*,
body: object,
response: httpx.Response,
) -> APIStatusError:
if response.status_code == 400:
return _exceptions.BadRequestError(err_msg, response=response, body=body)
if response.status_code == 401:
return _exceptions.AuthenticationError(err_msg, response=response, body=body)
if response.status_code == 403:
return _exceptions.PermissionDeniedError(err_msg, response=response, body=body)
if response.status_code == 404:
return _exceptions.NotFoundError(err_msg, response=response, body=body)
if response.status_code == 409:
return _exceptions.ConflictError(err_msg, response=response, body=body)
if response.status_code == 422:
return _exceptions.UnprocessableEntityError(err_msg, response=response, body=body)
if response.status_code == 429:
return _exceptions.RateLimitError(err_msg, response=response, body=body)
if response.status_code >= 500:
return _exceptions.InternalServerError(err_msg, response=response, body=body)
return APIStatusError(err_msg, response=response, body=body)
class AsyncReplicate(AsyncAPIClient):
# client options
bearer_token: str
def __init__(
self,
*,
bearer_token: str | None = None,
base_url: str | httpx.URL | None = None,
timeout: Union[float, Timeout, None, NotGiven] = NOT_GIVEN,
max_retries: int = DEFAULT_MAX_RETRIES,
default_headers: Mapping[str, str] | None = None,
default_query: Mapping[str, object] | None = None,
# Configure a custom httpx client.
# We provide a `DefaultAsyncHttpxClient` class that you can pass to retain the default values we use for `limits`, `timeout` & `follow_redirects`.
# See the [httpx documentation](https://www.python-httpx.org/api/#asyncclient) for more details.
http_client: httpx.AsyncClient | None = None,
# Enable or disable schema validation for data returned by the API.
# When enabled an error APIResponseValidationError is raised
# if the API responds with invalid data for the expected schema.
#
# This parameter may be removed or changed in the future.
# If you rely on this feature, please open a GitHub issue
# outlining your use-case to help us decide if it should be
# part of our public interface in the future.
_strict_response_validation: bool = False,
) -> None:
"""Construct a new async AsyncReplicate client instance.
This automatically infers the `bearer_token` argument from the `REPLICATE_API_TOKEN` environment variable if it is not provided.
"""
if bearer_token is None:
bearer_token = _get_api_token_from_environment()
if bearer_token is None:
raise ReplicateError(
"The bearer_token client option must be set either by passing bearer_token to the client or by setting the REPLICATE_API_TOKEN environment variable"
)
self.bearer_token = bearer_token
if base_url is None:
base_url = os.environ.get("REPLICATE_BASE_URL")
if base_url is None:
base_url = f"https://api.replicate.com/v1"
super().__init__(
version=__version__,
base_url=base_url,
max_retries=max_retries,
timeout=timeout,
http_client=http_client,
custom_headers=default_headers,
custom_query=default_query,
_strict_response_validation=_strict_response_validation,
)
@cached_property
def collections(self) -> AsyncCollectionsResource:
from .resources.collections import AsyncCollectionsResource
return AsyncCollectionsResource(self)
@cached_property
def deployments(self) -> AsyncDeploymentsResource:
from .resources.deployments import AsyncDeploymentsResource
return AsyncDeploymentsResource(self)
@cached_property
def hardware(self) -> AsyncHardwareResource:
from .resources.hardware import AsyncHardwareResource
return AsyncHardwareResource(self)
@cached_property
def account(self) -> AsyncAccountResource:
from .resources.account import AsyncAccountResource
return AsyncAccountResource(self)
@cached_property
def models(self) -> AsyncModelsResource:
from .resources.models import AsyncModelsResource
return AsyncModelsResource(self)
@cached_property
def predictions(self) -> AsyncPredictionsResource:
from .resources.predictions import AsyncPredictionsResource
return AsyncPredictionsResource(self)
@cached_property
def trainings(self) -> AsyncTrainingsResource:
from .resources.trainings import AsyncTrainingsResource
return AsyncTrainingsResource(self)
@cached_property
def webhooks(self) -> AsyncWebhooksResource:
from .resources.webhooks import AsyncWebhooksResource
return AsyncWebhooksResource(self)
@cached_property
def files(self) -> AsyncFilesResource:
from .resources.files import AsyncFilesResource
return AsyncFilesResource(self)
@cached_property
def with_raw_response(self) -> AsyncReplicateWithRawResponse:
return AsyncReplicateWithRawResponse(self)
@cached_property
def with_streaming_response(self) -> AsyncReplicateWithStreamedResponse:
return AsyncReplicateWithStreamedResponse(self)
@cached_property
def poll_interval(self) -> float:
return float(os.environ.get("REPLICATE_POLL_INTERVAL", "0.5"))
@property
@override
def qs(self) -> Querystring:
return Querystring(array_format="comma")
@property
@override
def auth_headers(self) -> dict[str, str]:
bearer_token = self.bearer_token
return {"Authorization": f"Bearer {bearer_token}"}
@property
@override
def default_headers(self) -> dict[str, str | Omit]:
return {
**super().default_headers,
"X-Stainless-Async": f"async:{get_async_library()}",
**self._custom_headers,
}
async def run(
self,
ref: Union[Model, Version, ModelVersionIdentifier, str],
*,
use_file_output: bool = True,
file_encoding_strategy: Optional["FileEncodingStrategy"] = None,
wait: Union[int, bool, NotGiven] = NOT_GIVEN,
**params: Unpack[PredictionCreateParamsWithoutVersion],
) -> Any:
"""
Run a model prediction asynchronously.
Args:
ref: Reference to the model or version to run. Can be:
- A string containing a version ID (e.g. "5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa")
- A string with owner/name format (e.g. "replicate/hello-world")
- A string with owner/name:version format (e.g. "replicate/hello-world:5c7d5dc6...")
- A Model instance with owner and name attributes
- A Version instance with id attribute
- A ModelVersionIdentifier dictionary with owner, name, and/or version keys
use_file_output: If True (default), convert output URLs to AsyncFileOutput objects
file_encoding_strategy: Strategy for encoding file inputs, options are "base64" or "url"
wait: If True (default), wait for the prediction to complete. If False, return immediately.
If an integer, wait up to that many seconds.
**params: Additional parameters to pass to the prediction creation endpoint including
the required "input" dictionary with model-specific parameters
Returns:
The prediction output, which could be a basic type (str, int, etc.), an AsyncFileOutput object,
a list of AsyncFileOutput objects, or a dictionary of AsyncFileOutput objects, depending on what
the model returns.
Raises:
ModelError: If the model run fails
ValueError: If the reference format is invalid
TypeError: If both wait and prefer parameters are provided
"""
from .lib._predictions_run import async_run
return await async_run(
self,
ref,
wait=wait,
use_file_output=use_file_output,
file_encoding_strategy=file_encoding_strategy,
**params,
)
@overload
def use(
self,
ref: Union[str, "FunctionRef[Input, Output]"],
*,
hint: Optional[Callable["Input", "Output"]] = None,
streaming: Literal[False] = False,
) -> "AsyncFunction[Input, Output]": ...
@overload
def use(
self,
ref: Union[str, "FunctionRef[Input, Output]"],
*,
hint: Optional[Callable["Input", "Output"]] = None,
streaming: Literal[True],
) -> "AsyncFunction[Input, AsyncIterator[Output]]": ...
def use(
self,
ref: Union[str, "FunctionRef[Input, Output]"],
*,
hint: Optional[Callable["Input", "Output"]] = None,
streaming: bool = False,
) -> Union["AsyncFunction[Input, Output]", "AsyncFunction[Input, AsyncIterator[Output]]"]:
"""
Use a Replicate model as an async function.
Example:
flux_dev = replicate.use("black-forest-labs/flux-dev", use_async=True)
output = await flux_dev(prompt="make me a sandwich")
"""
from .lib._predictions_use import use as _use
# TODO: Fix mypy overload matching for streaming parameter
return _use(self, ref, hint=hint, streaming=streaming) # type: ignore[call-overload, no-any-return]
def copy(
self,
*,
bearer_token: str | None = None,
base_url: str | httpx.URL | None = None,
timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
http_client: httpx.AsyncClient | None = None,
max_retries: int | NotGiven = NOT_GIVEN,
default_headers: Mapping[str, str] | None = None,
set_default_headers: Mapping[str, str] | None = None,
default_query: Mapping[str, object] | None = None,
set_default_query: Mapping[str, object] | None = None,
_extra_kwargs: Mapping[str, Any] = {},
) -> Self:
"""
Create a new client instance re-using the same options given to the current client with optional overriding.
"""
if default_headers is not None and set_default_headers is not None:
raise ValueError("The `default_headers` and `set_default_headers` arguments are mutually exclusive")
if default_query is not None and set_default_query is not None:
raise ValueError("The `default_query` and `set_default_query` arguments are mutually exclusive")
headers = self._custom_headers
if default_headers is not None:
headers = {**headers, **default_headers}
elif set_default_headers is not None:
headers = set_default_headers
params = self._custom_query
if default_query is not None:
params = {**params, **default_query}
elif set_default_query is not None:
params = set_default_query
http_client = http_client or self._client
return self.__class__(
bearer_token=bearer_token or self.bearer_token,
base_url=base_url or self.base_url,
timeout=self.timeout if isinstance(timeout, NotGiven) else timeout,
http_client=http_client,
max_retries=max_retries if is_given(max_retries) else self.max_retries,
default_headers=headers,
default_query=params,
**_extra_kwargs,
)
# Alias for `copy` for nicer inline usage, e.g.
# client.with_options(timeout=10).foo.create(...)
with_options = copy
async def search(
self,
*,
query: str,
limit: int | NotGiven = NOT_GIVEN,
# Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
# The extra values given here take precedence over values defined on the client or passed to this method.
extra_headers: Headers | None = None,
extra_query: Query | None = None,
extra_body: Body | None = None,
timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
) -> SearchResponse:
"""
Search for public models, collections, and docs using a text query.
For models, the response includes all model data, plus a new `metadata` object
with the following fields:
- `generated_description`: A longer and more detailed AI-generated description
of the model
- `tags`: An array of tags for the model
- `score`: A score for the model's relevance to the search query
Example cURL request:
```console
curl -s \\
-H "Authorization: Bearer $REPLICATE_API_TOKEN" \\
"https://api.replicate.com/v1/search?query=nano+banana"
```
Note: This search API is currently in beta and may change in future versions.
Args:
query: The search query string
limit: Maximum number of model results to return (1-50, defaults to 20)
extra_headers: Send extra headers
extra_query: Add additional query parameters to the request
extra_body: Add additional JSON properties to the request
timeout: Override the client-level default timeout for this request, in seconds
"""
return await self.get(
"/search",
options=make_request_options(
extra_headers=extra_headers,
extra_query=extra_query,
extra_body=extra_body,
timeout=timeout,
query=await async_maybe_transform(
{
"query": query,
"limit": limit,
},
client_search_params.ClientSearchParams,
),
),
cast_to=SearchResponse,
)
@override
def _make_status_error(
self,
err_msg: str,
*,
body: object,
response: httpx.Response,
) -> APIStatusError:
if response.status_code == 400:
return _exceptions.BadRequestError(err_msg, response=response, body=body)
if response.status_code == 401:
return _exceptions.AuthenticationError(err_msg, response=response, body=body)
if response.status_code == 403:
return _exceptions.PermissionDeniedError(err_msg, response=response, body=body)
if response.status_code == 404:
return _exceptions.NotFoundError(err_msg, response=response, body=body)
if response.status_code == 409:
return _exceptions.ConflictError(err_msg, response=response, body=body)
if response.status_code == 422:
return _exceptions.UnprocessableEntityError(err_msg, response=response, body=body)
if response.status_code == 429:
return _exceptions.RateLimitError(err_msg, response=response, body=body)
if response.status_code >= 500:
return _exceptions.InternalServerError(err_msg, response=response, body=body)
return APIStatusError(err_msg, response=response, body=body)
class ReplicateWithRawResponse:
_client: Replicate
def __init__(self, client: Replicate) -> None:
self._client = client
self.search = to_raw_response_wrapper(
client.search,
)
@cached_property
def collections(self) -> collections.CollectionsResourceWithRawResponse:
from .resources.collections import CollectionsResourceWithRawResponse
return CollectionsResourceWithRawResponse(self._client.collections)
@cached_property
def deployments(self) -> deployments.DeploymentsResourceWithRawResponse:
from .resources.deployments import DeploymentsResourceWithRawResponse
return DeploymentsResourceWithRawResponse(self._client.deployments)
@cached_property
def hardware(self) -> hardware.HardwareResourceWithRawResponse:
from .resources.hardware import HardwareResourceWithRawResponse
return HardwareResourceWithRawResponse(self._client.hardware)
@cached_property
def account(self) -> account.AccountResourceWithRawResponse:
from .resources.account import AccountResourceWithRawResponse
return AccountResourceWithRawResponse(self._client.account)
@cached_property
def models(self) -> models.ModelsResourceWithRawResponse:
from .resources.models import ModelsResourceWithRawResponse
return ModelsResourceWithRawResponse(self._client.models)
@cached_property
def predictions(self) -> predictions.PredictionsResourceWithRawResponse:
from .resources.predictions import PredictionsResourceWithRawResponse
return PredictionsResourceWithRawResponse(self._client.predictions)
@cached_property
def trainings(self) -> trainings.TrainingsResourceWithRawResponse:
from .resources.trainings import TrainingsResourceWithRawResponse
return TrainingsResourceWithRawResponse(self._client.trainings)
@cached_property
def webhooks(self) -> webhooks.WebhooksResourceWithRawResponse:
from .resources.webhooks import WebhooksResourceWithRawResponse
return WebhooksResourceWithRawResponse(self._client.webhooks)
@cached_property
def files(self) -> files.FilesResourceWithRawResponse:
from .resources.files import FilesResourceWithRawResponse
return FilesResourceWithRawResponse(self._client.files)
class AsyncReplicateWithRawResponse:
_client: AsyncReplicate
def __init__(self, client: AsyncReplicate) -> None:
self._client = client
self.search = async_to_raw_response_wrapper(
client.search,
)
@cached_property
def collections(self) -> collections.AsyncCollectionsResourceWithRawResponse:
from .resources.collections import AsyncCollectionsResourceWithRawResponse
return AsyncCollectionsResourceWithRawResponse(self._client.collections)
@cached_property
def deployments(self) -> deployments.AsyncDeploymentsResourceWithRawResponse:
from .resources.deployments import AsyncDeploymentsResourceWithRawResponse
return AsyncDeploymentsResourceWithRawResponse(self._client.deployments)
@cached_property
def hardware(self) -> hardware.AsyncHardwareResourceWithRawResponse:
from .resources.hardware import AsyncHardwareResourceWithRawResponse
return AsyncHardwareResourceWithRawResponse(self._client.hardware)
@cached_property
def account(self) -> account.AsyncAccountResourceWithRawResponse:
from .resources.account import AsyncAccountResourceWithRawResponse
return AsyncAccountResourceWithRawResponse(self._client.account)
@cached_property
def models(self) -> models.AsyncModelsResourceWithRawResponse:
from .resources.models import AsyncModelsResourceWithRawResponse
return AsyncModelsResourceWithRawResponse(self._client.models)
@cached_property
def predictions(self) -> predictions.AsyncPredictionsResourceWithRawResponse:
from .resources.predictions import AsyncPredictionsResourceWithRawResponse
return AsyncPredictionsResourceWithRawResponse(self._client.predictions)
@cached_property
def trainings(self) -> trainings.AsyncTrainingsResourceWithRawResponse:
from .resources.trainings import AsyncTrainingsResourceWithRawResponse
return AsyncTrainingsResourceWithRawResponse(self._client.trainings)
@cached_property
def webhooks(self) -> webhooks.AsyncWebhooksResourceWithRawResponse:
from .resources.webhooks import AsyncWebhooksResourceWithRawResponse
return AsyncWebhooksResourceWithRawResponse(self._client.webhooks)
@cached_property
def files(self) -> files.AsyncFilesResourceWithRawResponse:
from .resources.files import AsyncFilesResourceWithRawResponse
return AsyncFilesResourceWithRawResponse(self._client.files)
class ReplicateWithStreamedResponse:
_client: Replicate
def __init__(self, client: Replicate) -> None:
self._client = client
self.search = to_streamed_response_wrapper(
client.search,
)
@cached_property
def collections(self) -> collections.CollectionsResourceWithStreamingResponse:
from .resources.collections import CollectionsResourceWithStreamingResponse
return CollectionsResourceWithStreamingResponse(self._client.collections)
@cached_property
def deployments(self) -> deployments.DeploymentsResourceWithStreamingResponse:
from .resources.deployments import DeploymentsResourceWithStreamingResponse
return DeploymentsResourceWithStreamingResponse(self._client.deployments)
@cached_property
def hardware(self) -> hardware.HardwareResourceWithStreamingResponse: