fal 1.8.1__py3-none-any.whl → 1.8.2__py3-none-any.whl

fal/_fal_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '1.8.1'
-__version_tuple__ = version_tuple = (1, 8, 1)
+__version__ = version = '1.8.2'
+__version_tuple__ = version_tuple = (1, 8, 2)

fal/api.py

@@ -403,6 +403,7 @@ class FalServerlessHost(Host):
             "setup_function",
             "metadata",
             "request_timeout",
+            "startup_timeout",
             "_base_image",
             "_scheduler",
             "_scheduler_options",
@@ -452,6 +453,7 @@ class FalServerlessHost(Host):
         max_multiplexing = options.host.get("max_multiplexing")
         exposed_port = options.get_exposed_port()
         request_timeout = options.host.get("request_timeout")
+        startup_timeout = options.host.get("startup_timeout")
         machine_requirements = MachineRequirements(
             machine_types=machine_type,  # type: ignore
             num_gpus=options.host.get("num_gpus"),
@@ -464,6 +466,7 @@ class FalServerlessHost(Host):
             max_concurrency=max_concurrency,
             min_concurrency=min_concurrency,
             request_timeout=request_timeout,
+            startup_timeout=startup_timeout,
         )
 
         partial_func = _prepare_partial_func(func)
@@ -526,7 +529,7 @@ class FalServerlessHost(Host):
         exposed_port = options.get_exposed_port()
         setup_function = options.host.get("setup_function", None)
         request_timeout = options.host.get("request_timeout")
-
+        startup_timeout = options.host.get("startup_timeout")
         machine_requirements = MachineRequirements(
             machine_types=machine_type,  # type: ignore
             num_gpus=options.host.get("num_gpus"),
@@ -539,6 +542,7 @@ class FalServerlessHost(Host):
             max_concurrency=max_concurrency,
             min_concurrency=min_concurrency,
             request_timeout=request_timeout,
+            startup_timeout=startup_timeout,
         )
 
         return_value = _UNSET
@@ -705,6 +709,7 @@ def function(
     max_multiplexing: int = FAL_SERVERLESS_DEFAULT_MAX_MULTIPLEXING,
     min_concurrency: int = FAL_SERVERLESS_DEFAULT_MIN_CONCURRENCY,
     request_timeout: int | None = None,
+    startup_timeout: int | None = None,
     setup_function: Callable[..., None] | None = None,
     _base_image: str | None = None,
     _scheduler: str | None = None,
@@ -732,6 +737,7 @@ def function(
     max_multiplexing: int = FAL_SERVERLESS_DEFAULT_MAX_MULTIPLEXING,
     min_concurrency: int = FAL_SERVERLESS_DEFAULT_MIN_CONCURRENCY,
     request_timeout: int | None = None,
+    startup_timeout: int | None = None,
     setup_function: Callable[..., None] | None = None,
     _base_image: str | None = None,
     _scheduler: str | None = None,
@@ -809,6 +815,7 @@ def function(
     max_multiplexing: int = FAL_SERVERLESS_DEFAULT_MAX_MULTIPLEXING,
     min_concurrency: int = FAL_SERVERLESS_DEFAULT_MIN_CONCURRENCY,
     request_timeout: int | None = None,
+    startup_timeout: int | None = None,
     setup_function: Callable[..., None] | None = None,
     _base_image: str | None = None,
     _scheduler: str | None = None,
@@ -841,6 +848,7 @@ def function(
     max_multiplexing: int = FAL_SERVERLESS_DEFAULT_MAX_MULTIPLEXING,
     min_concurrency: int = FAL_SERVERLESS_DEFAULT_MIN_CONCURRENCY,
     request_timeout: int | None = None,
+    startup_timeout: int | None = None,
     setup_function: Callable[..., None] | None = None,
     _base_image: str | None = None,
     _scheduler: str | None = None,
@@ -867,6 +875,7 @@ def function(
     max_multiplexing: int = FAL_SERVERLESS_DEFAULT_MAX_MULTIPLEXING,
     min_concurrency: int = FAL_SERVERLESS_DEFAULT_MIN_CONCURRENCY,
     request_timeout: int | None = None,
+    startup_timeout: int | None = None,
     setup_function: Callable[..., None] | None = None,
     _base_image: str | None = None,
     _scheduler: str | None = None,
@@ -893,6 +902,7 @@ def function(
     max_multiplexing: int = FAL_SERVERLESS_DEFAULT_MAX_MULTIPLEXING,
     min_concurrency: int = FAL_SERVERLESS_DEFAULT_MIN_CONCURRENCY,
     request_timeout: int | None = None,
+    startup_timeout: int | None = None,
     setup_function: Callable[..., None] | None = None,
     _base_image: str | None = None,
     _scheduler: str | None = None,

fal/app.py

@@ -271,6 +271,7 @@ class App(fal.api.BaseServable):
     app_name: ClassVar[str]
     app_auth: ClassVar[Literal["private", "public", "shared"]] = "private"
     request_timeout: ClassVar[int | None] = None
+    startup_timeout: ClassVar[int | None] = None
 
     isolate_channel: async_grpc.Channel | None = None
 
@@ -282,6 +283,9 @@ class App(fal.api.BaseServable):
         if cls.request_timeout is not None:
             cls.host_kwargs["request_timeout"] = cls.request_timeout
 
+        if cls.startup_timeout is not None:
+            cls.host_kwargs["startup_timeout"] = cls.startup_timeout
+
         cls.app_name = getattr(cls, "app_name", app_name)
 
         if cls.__init__ is not App.__init__:

fal/cli/apps.py

@@ -19,6 +19,9 @@ def _apps_table(apps: list[AliasInfo]):
     table.add_column("Max Concurrency")
     table.add_column("Max Multiplexing")
     table.add_column("Keep Alive")
+    table.add_column("Request Timeout")
+    table.add_column("Startup Timeout")
+    table.add_column("Machine Type")
     table.add_column("Active Workers")
 
     for app in apps:
@@ -30,6 +33,9 @@ def _apps_table(apps: list[AliasInfo]):
             str(app.max_concurrency),
             str(app.max_multiplexing),
             str(app.keep_alive),
+            str(app.request_timeout),
+            str(app.startup_timeout),
+            " ".join(app.machine_types),
             str(app.active_runners),
         )
 
@@ -67,6 +73,9 @@ def _app_rev_table(revs: list[ApplicationInfo]):
     table.add_column("Max Concurrency")
     table.add_column("Max Multiplexing")
     table.add_column("Keep Alive")
+    table.add_column("Request Timeout")
+    table.add_column("Startup Timeout")
+    table.add_column("Machine Type")
     table.add_column("Active Workers")
 
     for rev in revs:
@@ -76,6 +85,9 @@ def _app_rev_table(revs: list[ApplicationInfo]):
             str(rev.max_concurrency),
             str(rev.max_multiplexing),
             str(rev.keep_alive),
+            str(rev.request_timeout),
+            str(rev.startup_timeout),
+            " ".join(rev.machine_types),
             str(rev.active_runners),
         )
 
@@ -114,6 +126,8 @@ def _scale(args):
             and args.max_multiplexing is None
             and args.max_concurrency is None
             and args.min_concurrency is None
+            and args.request_timeout is None
+            and args.startup_timeout is None
         ):
             args.console.log("No parameters for update were provided, ignoring.")
             return
@@ -124,6 +138,8 @@ def _scale(args):
             max_multiplexing=args.max_multiplexing,
             max_concurrency=args.max_concurrency,
             min_concurrency=args.min_concurrency,
+            request_timeout=args.request_timeout,
+            startup_timeout=args.startup_timeout,
         )
         table = _apps_table([alias_info])
 
@@ -162,6 +178,16 @@ def _add_scale_parser(subparsers, parents):
         type=int,
         help="Minimum concurrency",
     )
+    parser.add_argument(
+        "--request-timeout",
+        type=int,
+        help="Request timeout (seconds).",
+    )
+    parser.add_argument(
+        "--startup-timeout",
+        type=int,
+        help="Startup timeout (seconds).",
+    )
     parser.set_defaults(func=_scale)
 
 

fal/cli/deploy.py

@@ -111,10 +111,16 @@ def _deploy_from_reference(
     )
 
     if app_id:
-        gateway_host = _remove_http_and_port_from_url(host.url)
-        gateway_host = (
-            gateway_host.replace("api.", "").replace("alpha.", "").replace("ai", "run")
-        )
+        env_host = _remove_http_and_port_from_url(host.url)
+        env_host = env_host.replace("api.", "").replace("alpha.", "")
+
+        env_host_parts = env_host.split(".")
+
+        # keep the last 3 parts
+        playground_host = ".".join(env_host_parts[-3:])
+
+        # just replace .ai for .run
+        endpoint_host = env_host.replace(".ai", ".run")
 
         args.console.print(
             "Registered a new revision for function "
@@ -123,12 +129,12 @@ def _deploy_from_reference(
         args.console.print("Playground:")
         for endpoint in loaded.endpoints:
             args.console.print(
-                f"\thttps://fal.ai/models/{user.username}/{app_name}{endpoint}"
+                f"\thttps://{playground_host}/models/{user.username}/{app_name}{endpoint}"
             )
         args.console.print("Endpoints:")
         for endpoint in loaded.endpoints:
             args.console.print(
-                f"\thttps://{gateway_host}/{user.username}/{app_name}{endpoint}"
+                f"\thttps://{endpoint_host}/{user.username}/{app_name}{endpoint}"
             )
 
 

fal/sdk.py

@@ -196,6 +196,9 @@ class ApplicationInfo:
     max_multiplexing: int
     active_runners: int
     min_concurrency: int
+    machine_types: list[str]
+    request_timeout: int
+    startup_timeout: int
 
 
 @dataclass
@@ -208,6 +211,9 @@ class AliasInfo:
     max_multiplexing: int
     active_runners: int
     min_concurrency: int
+    machine_types: list[str]
+    request_timeout: int
+    startup_timeout: int
 
 
 @dataclass
@@ -313,6 +319,9 @@ def _from_grpc_application_info(
         max_multiplexing=message.max_multiplexing,
         active_runners=message.active_runners,
         min_concurrency=message.min_concurrency,
+        machine_types=message.machine_types,
+        request_timeout=message.request_timeout,
+        startup_timeout=message.startup_timeout,
     )
 
 
@@ -336,6 +345,9 @@ def _from_grpc_alias_info(message: isolate_proto.AliasInfo) -> AliasInfo:
         max_multiplexing=message.max_multiplexing,
         active_runners=message.active_runners,
         min_concurrency=message.min_concurrency,
+        machine_types=message.machine_types,
+        request_timeout=message.request_timeout,
+        startup_timeout=message.startup_timeout,
     )
 
 
@@ -402,6 +414,7 @@ class MachineRequirements:
     max_multiplexing: int | None = None
     min_concurrency: int | None = None
     request_timeout: int | None = None
+    startup_timeout: int | None = None
 
     def __post_init__(self):
         if isinstance(self.machine_types, str):
@@ -519,6 +532,7 @@ class FalServerlessConnection:
                 min_concurrency=machine_requirements.min_concurrency,
                 max_multiplexing=machine_requirements.max_multiplexing,
                 request_timeout=machine_requirements.request_timeout,
+                startup_timeout=machine_requirements.startup_timeout,
             )
         else:
             wrapped_requirements = None
@@ -562,6 +576,8 @@ class FalServerlessConnection:
         max_multiplexing: int | None = None,
         max_concurrency: int | None = None,
         min_concurrency: int | None = None,
+        request_timeout: int | None = None,
+        startup_timeout: int | None = None,
     ) -> AliasInfo:
         request = isolate_proto.UpdateApplicationRequest(
             application_name=application_name,
@@ -569,6 +585,8 @@ class FalServerlessConnection:
             max_multiplexing=max_multiplexing,
             max_concurrency=max_concurrency,
             min_concurrency=min_concurrency,
+            request_timeout=request_timeout,
+            startup_timeout=startup_timeout,
         )
         res: isolate_proto.UpdateApplicationResult = self.stub.UpdateApplication(
             request
@@ -614,6 +632,7 @@ class FalServerlessConnection:
                 max_multiplexing=machine_requirements.max_multiplexing,
                 min_concurrency=machine_requirements.min_concurrency,
                 request_timeout=machine_requirements.request_timeout,
+                startup_timeout=machine_requirements.startup_timeout,
             )
         else:
             wrapped_requirements = None

fal/toolkit/file/file.py

@@ -59,7 +59,7 @@ def get_builtin_repository(id: RepositoryId | FileRepository) -> FileRepository:
 get_builtin_repository.__module__ = "__main__"
 
 DEFAULT_REPOSITORY: FileRepository | RepositoryId = "fal_v3"
-FALLBACK_REPOSITORY: FileRepository | RepositoryId = "cdn"
+FALLBACK_REPOSITORY: FileRepository | RepositoryId = "fal"
 OBJECT_LIFECYCLE_PREFERENCE_KEY = "x-fal-object-lifecycle-preference"
 
 
@@ -153,9 +153,12 @@ class File(BaseModel):
         object_lifecycle_preference = (
             request_lifecycle_preference(request) or LIFECYCLE_PREFERENCE.get()
         )
+        save_kwargs.setdefault(
+            "object_lifecycle_preference", object_lifecycle_preference
+        )
 
         try:
-            url = repo.save(fdata, object_lifecycle_preference, **save_kwargs)
+            url = repo.save(fdata, **save_kwargs)
         except Exception:
             if not fallback_repository:
                 raise

{fal-1.8.1.dist-info → fal-1.8.2.dist-info}/METADATA

@@ -1,12 +1,12 @@
 Metadata-Version: 2.2
 Name: fal
-Version: 1.8.1
+Version: 1.8.2
 Summary: fal is an easy-to-use Serverless Python Framework
 Author: Features & Labels <support@fal.ai>
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 Requires-Dist: isolate[build]<0.16.0,>=0.15.0
-Requires-Dist: isolate-proto<0.7.0,>=0.6.0
+Requires-Dist: isolate-proto<0.7.0,>=0.6.6
 Requires-Dist: grpcio==1.64.0
 Requires-Dist: dill==0.3.7
 Requires-Dist: cloudpickle==3.0.0

{fal-1.8.1.dist-info → fal-1.8.2.dist-info}/RECORD

@@ -1,10 +1,10 @@
 fal/__init__.py,sha256=wXs1G0gSc7ZK60-bHe-B2m0l_sA6TrFk4BxY0tMoLe8,784
 fal/__main__.py,sha256=4JMK66Wj4uLZTKbF-sT3LAxOsr6buig77PmOkJCRRxw,83
-fal/_fal_version.py,sha256=0CBOWvcgj3z82UQGmgPZNYucWtZfu-kClLlOblR1VJs,411
+fal/_fal_version.py,sha256=wyeOGWTUT7rTWqRZmS_XmHAtprBx1jO6L7FxPeYTyMM,411
 fal/_serialization.py,sha256=rD2YiSa8iuzCaZohZwN_MPEB-PpSKbWRDeaIDpTEjyY,7653
 fal/_version.py,sha256=EBGqrknaf1WygENX-H4fBefLvHryvJBBGtVJetaB0NY,266
-fal/api.py,sha256=TJNuez-xjzHfIAA7ZLmEMksdB3U5EM7bl16ooRL_1Ak,43499
-fal/app.py,sha256=0cm7wZXdusZXyV9nJmlFJl-7Jgxir6954OaO9Lj2ITk,23178
+fal/api.py,sha256=ZbUoe12y6sVg5-bPbzFtqHTqZoQonVDaZRjlGNaHbcs,43983
+fal/app.py,sha256=hk3rT-3fZY7jsgK270JPC26Wnc2GhhtrDcjRCUQRr6Q,23341
 fal/apps.py,sha256=RpmElElJnDYjsTRQOdNYiJwd74GEOGYA38L5O5GzNEg,11068
 fal/config.py,sha256=hgI3kW4_2NoFsrYEiPss0mnDTr8_Td2z0pVgm93wi9o,600
 fal/container.py,sha256=9XslBET-NCG2V3-Wmof8c7eHrRoxCye88Ym7CskqCk0,1639
@@ -12,7 +12,7 @@ fal/files.py,sha256=QgfYfMKmNobMPufrAP_ga1FKcIAlSbw18Iar1-0qepo,2650
 fal/flags.py,sha256=oWN_eidSUOcE9wdPK_77si3A1fpgOC0UEERPsvNLIMc,842
 fal/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 fal/rest_client.py,sha256=kGBGmuyHfX1lR910EoKCYPjsyU8MdXawT_cW2q8Sajc,568
-fal/sdk.py,sha256=HjlToPJkG0Z5h_D0D2FK43i3JFKeO4r2IhCGx4B82Z8,22564
+fal/sdk.py,sha256=p-D_4b7wDpASrMXIeni-7LgFbQ8nUG2tvPn9Z4QDqos,23365
 fal/sync.py,sha256=ZuIJA2-hTPNANG9B_NNJZUsO68EIdTH0dc9MzeVE2VU,4340
 fal/utils.py,sha256=9q_QrQBlQN3nZYA1kEGRfhJWi4RjnO4H1uQswfaei9w,2146
 fal/workflows.py,sha256=Zl4f6Bs085hY40zmqScxDUyCu7zXkukDbW02iYOLTTI,14805
@@ -21,11 +21,11 @@ fal/auth/auth0.py,sha256=rSG1mgH-QGyKfzd7XyAaj1AYsWt-ho8Y_LZ-FUVWzh4,5421
 fal/auth/local.py,sha256=sndkM6vKpeVny6NHTacVlTbiIFqaksOmw0Viqs_RN1U,1790
 fal/cli/__init__.py,sha256=padK4o0BFqq61kxAA1qQ0jYr2SuhA2mf90B3AaRkmJA,37
 fal/cli/_utils.py,sha256=45G0LEz2bW-69MUQKPdatVE_CBC2644gC-V0qdNEsco,1252
-fal/cli/apps.py,sha256=Fo4iUpd6FGTUcIp22WcssE1CaEn_BLKzK_E4JPsXhVI,8179
+fal/cli/apps.py,sha256=DeNzouQSv2cFZKDF4A5xdl3qzQ6NWbbgF3pivgmaIiI,9079
 fal/cli/auth.py,sha256=--MhfHGwxmtHbRkGioyn1prKn_U-pBzbz0G_QeZou-U,1352
 fal/cli/create.py,sha256=a8WDq-nJLFTeoIXqpb5cr7GR7YR9ZZrQCawNm34KXXE,627
 fal/cli/debug.py,sha256=u_urnyFzSlNnrq93zz_GXE9FX4VyVxDoamJJyrZpFI0,1312
-fal/cli/deploy.py,sha256=-woTZObzntUenPFmWJwDaeCmBl3Vb7jqSkhPCIfk2SM,7581
+fal/cli/deploy.py,sha256=Wu8wxR72od2GAp1OF-FYO6WSWt4umaHysL3HrO_bzEo,7764
 fal/cli/doctor.py,sha256=U4ne9LX5gQwNblsYQ27XdO8AYDgbYjTO39EtxhwexRM,983
 fal/cli/keys.py,sha256=trDpA3LJu9S27qE_K8Hr6fKLK4vwVzbxUHq8TFrV4pw,3157
 fal/cli/main.py,sha256=O0i9wdLPxcd1u4CvXit-ufkT_UnON-baTN6v9HaHPmw,2027
@@ -50,7 +50,7 @@ fal/toolkit/exceptions.py,sha256=elHZ7dHCJG5zlHGSBbz-ilkZe9QUvQMomJFi8Pt91LA,198
 fal/toolkit/optimize.py,sha256=p75sovF0SmRP6zxzpIaaOmqlxvXB_xEz3XPNf59EF7w,1339
 fal/toolkit/types.py,sha256=kkbOsDKj1qPGb1UARTBp7yuJ5JUuyy7XQurYUBCdti8,4064
 fal/toolkit/file/__init__.py,sha256=FbNl6wD-P0aSSTUwzHt4HujBXrbC3ABmaigPQA4hRfg,70
-fal/toolkit/file/file.py,sha256=-gccCKnarTu6Nfm_0yQ0sJM9aadB5tUNvKS1PTqxiFc,9071
+fal/toolkit/file/file.py,sha256=ccvbWpfbvfXv11QKUEajjOze78RC7JOQDs7bZzZgPhw,9155
 fal/toolkit/file/types.py,sha256=MjZ6xAhKPv4rowLo2Vcbho0sX7AQ3lm3KFyYDcw0dL4,1845
 fal/toolkit/file/providers/fal.py,sha256=7JWTFXvAbtqakCIlA5gfKI8qU1umlWgWhvU5cXqzGVQ,36050
 fal/toolkit/file/providers/gcp.py,sha256=iQtkoYUqbmKKpC5srVOYtrruZ3reGRm5lz4kM8bshgk,2247
@@ -130,8 +130,8 @@ openapi_fal_rest/models/workflow_node_type.py,sha256=-FzyeY2bxcNmizKbJI8joG7byRi
 openapi_fal_rest/models/workflow_schema.py,sha256=4K5gsv9u9pxx2ItkffoyHeNjBBYf6ur5bN4m_zePZNY,2019
 openapi_fal_rest/models/workflow_schema_input.py,sha256=2OkOXWHTNsCXHWS6EGDFzcJKkW5FIap-2gfO233EvZQ,1191
 openapi_fal_rest/models/workflow_schema_output.py,sha256=EblwSPAGfWfYVWw_WSSaBzQVju296is9o28rMBAd0mc,1196
-fal-1.8.1.dist-info/METADATA,sha256=bEcIv3BFgkvrKOhJA_AgmQQkIQmKBdQq9oumzE0kmDA,4002
-fal-1.8.1.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-fal-1.8.1.dist-info/entry_points.txt,sha256=32zwTUC1U1E7nSTIGCoANQOQ3I7-qHG5wI6gsVz5pNU,37
-fal-1.8.1.dist-info/top_level.txt,sha256=r257X1L57oJL8_lM0tRrfGuXFwm66i1huwQygbpLmHw,21
-fal-1.8.1.dist-info/RECORD,,
+fal-1.8.2.dist-info/METADATA,sha256=l05QPJlpFqBqbK74aj6LCFFIxZbg7izsrg-LqS7ywXc,4002
+fal-1.8.2.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+fal-1.8.2.dist-info/entry_points.txt,sha256=32zwTUC1U1E7nSTIGCoANQOQ3I7-qHG5wI6gsVz5pNU,37
+fal-1.8.2.dist-info/top_level.txt,sha256=r257X1L57oJL8_lM0tRrfGuXFwm66i1huwQygbpLmHw,21
+fal-1.8.2.dist-info/RECORD,,