together 2.0.0a15__py3-none-any.whl → 2.0.0a16__py3-none-any.whl

together/lib/cli/api/{models.py → models/upload.py}

@@ -1,63 +1,15 @@
 import json as json_lib
-from typing import Any, Dict, List, Literal, Optional
+from typing import Literal, Optional
 
 import click
-from tabulate import tabulate
 
 from together import Together, omit
-from together._models import BaseModel
 from together._response import APIResponse as APIResponse
+from together.lib.cli.api._utils import handle_api_errors
 from together.types.model_upload_response import ModelUploadResponse
 
 
-@click.group()
-@click.pass_context
-def models(ctx: click.Context) -> None:
-    """Models API commands"""
-    pass
-
-
-@models.command()
-@click.option(
-    "--type",
-    type=click.Choice(["dedicated"]),
-    help="Filter models by type (dedicated: models that can be deployed as dedicated endpoints)",
-)
-@click.option(
-    "--json",
-    is_flag=True,
-    help="Output in JSON format",
-)
-@click.pass_context
-def list(ctx: click.Context, type: Optional[str], json: bool) -> None:
-    """List models"""
-    client: Together = ctx.obj
-
-    models_list = client.models.list(dedicated=type == "dedicated" if type else omit)
-
-    display_list: List[Dict[str, Any]] = []
-    model: BaseModel
-    for model in models_list:
-        display_list.append(
-            {
-                "ID": model.id,
-                "Name": model.display_name,
-                "Organization": model.organization,
-                "Type": model.type,
-                "Context Length": model.context_length,
-                "License": model.license,
-                "Input per 1M token": model.pricing.input if model.pricing else None,
-                "Output per 1M token": model.pricing.output if model.pricing else None,
-            }
-        )
-
-    if json:
-        click.echo(json_lib.dumps(display_list, indent=2))
-    else:
-        click.echo(tabulate(display_list, headers="keys", tablefmt="plain"))
-
-
-@models.command()
+@click.command()
 @click.option(
     "--model-name",
     required=True,
@@ -96,6 +48,7 @@ def list(ctx: click.Context, type: Optional[str], json: bool) -> None:
     help="Output in JSON format",
 )
 @click.pass_context
+@handle_api_errors("Models")
 def upload(
     ctx: click.Context,
     model_name: str,

together/resources/beta/clusters/clusters.py

@@ -28,9 +28,7 @@ from ....types.beta import cluster_create_params, cluster_update_params
 from ...._base_client import make_request_options
 from ....types.beta.cluster import Cluster
 from ....types.beta.cluster_list_response import ClusterListResponse
-from ....types.beta.cluster_create_response import ClusterCreateResponse
 from ....types.beta.cluster_delete_response import ClusterDeleteResponse
-from ....types.beta.cluster_update_response import ClusterUpdateResponse
 from ....types.beta.cluster_list_regions_response import ClusterListRegionsResponse
 
 __all__ = ["ClustersResource", "AsyncClustersResource"]
@@ -66,11 +64,11 @@ class ClustersResource(SyncAPIResource):
         billing_type: Literal["RESERVED", "ON_DEMAND"],
         cluster_name: str,
         driver_version: Literal["CUDA_12_5_555", "CUDA_12_6_560", "CUDA_12_6_565", "CUDA_12_8_570"],
-        duration_days: int,
         gpu_type: Literal["H100_SXM", "H200_SXM", "RTX_6000_PCI", "L40_PCIE", "B200_SXM", "H100_SXM_INF"],
         num_gpus: int,
         region: Literal["us-central-8", "us-central-4"],
         cluster_type: Literal["KUBERNETES", "SLURM"] | Omit = omit,
+        duration_days: int | Omit = omit,
         shared_volume: cluster_create_params.SharedVolume | Omit = omit,
         volume_id: str | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -79,17 +77,20 @@ class ClustersResource(SyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
-    ) -> ClusterCreateResponse:
-        """
-        Create GPU Cluster
+    ) -> Cluster:
+        """Create an Instant Cluster on Together's high-performance GPU clusters.
+
+        With
+        features like on-demand scaling, long-lived resizable high-bandwidth shared
+        DC-local storage, Kubernetes and Slurm cluster flavors, a REST API, and
+        Terraform support, you can run workloads flexibly without complex infrastructure
+        management.
 
         Args:
           cluster_name: Name of the GPU cluster.
 
           driver_version: NVIDIA driver version to use in the cluster.
 
-          duration_days: Duration in days to keep the cluster running.
-
           gpu_type: Type of GPU to use in the cluster
 
           num_gpus: Number of GPUs to allocate in the cluster. This must be multiple of 8. For
@@ -98,6 +99,8 @@ class ClustersResource(SyncAPIResource):
           region: Region to create the GPU cluster in. Valid values are us-central-8 and
               us-central-4.
 
+          duration_days: Duration in days to keep the cluster running.
+
           extra_headers: Send extra headers
 
           extra_query: Add additional query parameters to the request
@@ -113,11 +116,11 @@ class ClustersResource(SyncAPIResource):
                     "billing_type": billing_type,
                     "cluster_name": cluster_name,
                     "driver_version": driver_version,
-                    "duration_days": duration_days,
                     "gpu_type": gpu_type,
                     "num_gpus": num_gpus,
                     "region": region,
                     "cluster_type": cluster_type,
+                    "duration_days": duration_days,
                     "shared_volume": shared_volume,
                     "volume_id": volume_id,
                 },
@@ -126,7 +129,7 @@ class ClustersResource(SyncAPIResource):
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=ClusterCreateResponse,
+            cast_to=Cluster,
         )
 
     def retrieve(
@@ -141,7 +144,7 @@ class ClustersResource(SyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> Cluster:
         """
-        Get GPU cluster by cluster ID
+        Retrieve information about a specific GPU cluster.
 
         Args:
           extra_headers: Send extra headers
@@ -174,9 +177,9 @@ class ClustersResource(SyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
-    ) -> ClusterUpdateResponse:
+    ) -> Cluster:
         """
-        Update a GPU Cluster.
+        Update the configuration of an existing GPU cluster.
 
         Args:
           extra_headers: Send extra headers
@@ -201,7 +204,7 @@ class ClustersResource(SyncAPIResource):
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=ClusterUpdateResponse,
+            cast_to=Cluster,
         )
 
     def list(
@@ -235,7 +238,7 @@ class ClustersResource(SyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> ClusterDeleteResponse:
         """
-        Delete GPU cluster by cluster ID
+        Delete a GPU cluster by cluster ID.
 
         Args:
           extra_headers: Send extra headers
@@ -306,11 +309,11 @@ class AsyncClustersResource(AsyncAPIResource):
         billing_type: Literal["RESERVED", "ON_DEMAND"],
         cluster_name: str,
         driver_version: Literal["CUDA_12_5_555", "CUDA_12_6_560", "CUDA_12_6_565", "CUDA_12_8_570"],
-        duration_days: int,
         gpu_type: Literal["H100_SXM", "H200_SXM", "RTX_6000_PCI", "L40_PCIE", "B200_SXM", "H100_SXM_INF"],
         num_gpus: int,
         region: Literal["us-central-8", "us-central-4"],
         cluster_type: Literal["KUBERNETES", "SLURM"] | Omit = omit,
+        duration_days: int | Omit = omit,
         shared_volume: cluster_create_params.SharedVolume | Omit = omit,
         volume_id: str | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -319,17 +322,20 @@ class AsyncClustersResource(AsyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
-    ) -> ClusterCreateResponse:
-        """
-        Create GPU Cluster
+    ) -> Cluster:
+        """Create an Instant Cluster on Together's high-performance GPU clusters.
+
+        With
+        features like on-demand scaling, long-lived resizable high-bandwidth shared
+        DC-local storage, Kubernetes and Slurm cluster flavors, a REST API, and
+        Terraform support, you can run workloads flexibly without complex infrastructure
+        management.
 
         Args:
           cluster_name: Name of the GPU cluster.
 
           driver_version: NVIDIA driver version to use in the cluster.
 
-          duration_days: Duration in days to keep the cluster running.
-
           gpu_type: Type of GPU to use in the cluster
 
           num_gpus: Number of GPUs to allocate in the cluster. This must be multiple of 8. For
@@ -338,6 +344,8 @@ class AsyncClustersResource(AsyncAPIResource):
           region: Region to create the GPU cluster in. Valid values are us-central-8 and
               us-central-4.
 
+          duration_days: Duration in days to keep the cluster running.
+
           extra_headers: Send extra headers
 
           extra_query: Add additional query parameters to the request
@@ -353,11 +361,11 @@ class AsyncClustersResource(AsyncAPIResource):
                     "billing_type": billing_type,
                     "cluster_name": cluster_name,
                     "driver_version": driver_version,
-                    "duration_days": duration_days,
                     "gpu_type": gpu_type,
                     "num_gpus": num_gpus,
                     "region": region,
                     "cluster_type": cluster_type,
+                    "duration_days": duration_days,
                     "shared_volume": shared_volume,
                     "volume_id": volume_id,
                 },
@@ -366,7 +374,7 @@ class AsyncClustersResource(AsyncAPIResource):
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=ClusterCreateResponse,
+            cast_to=Cluster,
         )
 
     async def retrieve(
@@ -381,7 +389,7 @@ class AsyncClustersResource(AsyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> Cluster:
         """
-        Get GPU cluster by cluster ID
+        Retrieve information about a specific GPU cluster.
 
         Args:
           extra_headers: Send extra headers
@@ -414,9 +422,9 @@ class AsyncClustersResource(AsyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
-    ) -> ClusterUpdateResponse:
+    ) -> Cluster:
         """
-        Update a GPU Cluster.
+        Update the configuration of an existing GPU cluster.
 
         Args:
           extra_headers: Send extra headers
@@ -441,7 +449,7 @@ class AsyncClustersResource(AsyncAPIResource):
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=ClusterUpdateResponse,
+            cast_to=Cluster,
         )
 
     async def list(
@@ -475,7 +483,7 @@ class AsyncClustersResource(AsyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> ClusterDeleteResponse:
         """
-        Delete GPU cluster by cluster ID
+        Delete a GPU cluster by cluster ID.
 
         Args:
           extra_headers: Send extra headers

together/resources/beta/clusters/storage.py

@@ -18,7 +18,6 @@ from ...._base_client import make_request_options
 from ....types.beta.clusters import storage_create_params, storage_update_params
 from ....types.beta.clusters.cluster_storage import ClusterStorage
 from ....types.beta.clusters.storage_list_response import StorageListResponse
-from ....types.beta.clusters.storage_create_response import StorageCreateResponse
 from ....types.beta.clusters.storage_delete_response import StorageDeleteResponse
 
 __all__ = ["StorageResource", "AsyncStorageResource"]
@@ -56,13 +55,16 @@ class StorageResource(SyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
-    ) -> StorageCreateResponse:
-        """Create a shared volume.
+    ) -> ClusterStorage:
+        """
+        Instant Clusters supports long-lived, resizable in-DC shared storage with user
+        data persistence. You can dynamically create and attach volumes to your cluster
+        at cluster creation time, and resize as your data grows. All shared storage is
+        backed by multi-NIC bare metal paths, ensuring high-throughput and low-latency
+        performance for shared storage.
 
         Args:
-          region: Region name.
-
-        Usable regions can be found from `client.clusters.list_regions()`
+          region: Region name. Usable regions can be found from `client.clusters.list_regions()`
 
           size_tib: Volume size in whole tebibytes (TiB).
 
@@ -87,7 +89,7 @@ class StorageResource(SyncAPIResource):
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=StorageCreateResponse,
+            cast_to=ClusterStorage,
         )
 
     def retrieve(
@@ -102,7 +104,7 @@ class StorageResource(SyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> ClusterStorage:
         """
-        Get shared volume by volume Id.
+        Retrieve information about a specific shared volume.
 
         Args:
           extra_headers: Send extra headers
@@ -136,7 +138,7 @@ class StorageResource(SyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> ClusterStorage:
         """
-        Update a shared volume.
+        Update the configuration of an existing shared volume.
 
         Args:
           extra_headers: Send extra headers
@@ -192,8 +194,10 @@ class StorageResource(SyncAPIResource):
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> StorageDeleteResponse:
-        """
-        Delete shared volume by volume id.
+        """Delete a shared volume.
+
+        Note that if this volume is attached to a cluster,
+        deleting will fail.
 
         Args:
           extra_headers: Send extra headers
@@ -247,13 +251,16 @@ class AsyncStorageResource(AsyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
-    ) -> StorageCreateResponse:
-        """Create a shared volume.
+    ) -> ClusterStorage:
+        """
+        Instant Clusters supports long-lived, resizable in-DC shared storage with user
+        data persistence. You can dynamically create and attach volumes to your cluster
+        at cluster creation time, and resize as your data grows. All shared storage is
+        backed by multi-NIC bare metal paths, ensuring high-throughput and low-latency
+        performance for shared storage.
 
         Args:
-          region: Region name.
-
-        Usable regions can be found from `client.clusters.list_regions()`
+          region: Region name. Usable regions can be found from `client.clusters.list_regions()`
 
           size_tib: Volume size in whole tebibytes (TiB).
 
@@ -278,7 +285,7 @@ class AsyncStorageResource(AsyncAPIResource):
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=StorageCreateResponse,
+            cast_to=ClusterStorage,
         )
 
     async def retrieve(
@@ -293,7 +300,7 @@ class AsyncStorageResource(AsyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> ClusterStorage:
         """
-        Get shared volume by volume Id.
+        Retrieve information about a specific shared volume.
 
         Args:
           extra_headers: Send extra headers
@@ -327,7 +334,7 @@ class AsyncStorageResource(AsyncAPIResource):
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> ClusterStorage:
         """
-        Update a shared volume.
+        Update the configuration of an existing shared volume.
 
         Args:
           extra_headers: Send extra headers
@@ -383,8 +390,10 @@ class AsyncStorageResource(AsyncAPIResource):
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> StorageDeleteResponse:
-        """
-        Delete shared volume by volume id.
+        """Delete a shared volume.
+
+        Note that if this volume is attached to a cluster,
+        deleting will fail.
 
         Args:
           extra_headers: Send extra headers

together/types/__init__.py

@@ -1,6 +1,6 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from __future__ import annotations # noqa
+from __future__ import annotations  # noqa
 
 from .batch_job import BatchJob as BatchJob
 from .embedding import Embedding as Embedding
@@ -70,12 +70,13 @@ from .fine_tuning_list_checkpoints_response import (
 )
 
 # Manually added to minimize breaking changes from V1
-from .chat.chat_completion import ChatCompletion 
+from .chat.chat_completion import ChatCompletion
 from .chat.chat_completion_chunk import ChatCompletionChunk as ChatCompletionChunk
 from .chat.chat_completion_usage import ChatCompletionUsage
+
 UsageData = ChatCompletionUsage
 ChatCompletionResponse = ChatCompletion
 CompletionResponse = Completion
 ListEndpoint = EndpointListResponse
 ImageRequest = ImageGenerateParams
-ImageResponse = ImageFile
+ImageResponse = ImageFile

together/types/beta/__init__.py

@@ -6,7 +6,5 @@ from .cluster import Cluster as Cluster
 from .cluster_create_params import ClusterCreateParams as ClusterCreateParams
 from .cluster_list_response import ClusterListResponse as ClusterListResponse
 from .cluster_update_params import ClusterUpdateParams as ClusterUpdateParams
-from .cluster_create_response import ClusterCreateResponse as ClusterCreateResponse
 from .cluster_delete_response import ClusterDeleteResponse as ClusterDeleteResponse
-from .cluster_update_response import ClusterUpdateResponse as ClusterUpdateResponse
 from .cluster_list_regions_response import ClusterListRegionsResponse as ClusterListRegionsResponse

together/types/beta/cluster_create_params.py

@@ -16,9 +16,6 @@ class ClusterCreateParams(TypedDict, total=False):
     driver_version: Required[Literal["CUDA_12_5_555", "CUDA_12_6_560", "CUDA_12_6_565", "CUDA_12_8_570"]]
     """NVIDIA driver version to use in the cluster."""
 
-    duration_days: Required[int]
-    """Duration in days to keep the cluster running."""
-
     gpu_type: Required[Literal["H100_SXM", "H200_SXM", "RTX_6000_PCI", "L40_PCIE", "B200_SXM", "H100_SXM_INF"]]
     """Type of GPU to use in the cluster"""
 
@@ -36,6 +33,9 @@ class ClusterCreateParams(TypedDict, total=False):
 
     cluster_type: Literal["KUBERNETES", "SLURM"]
 
+    duration_days: int
+    """Duration in days to keep the cluster running."""
+
     shared_volume: SharedVolume
 
     volume_id: str

together/types/beta/clusters/__init__.py

@@ -6,5 +6,4 @@ from .cluster_storage import ClusterStorage as ClusterStorage
 from .storage_create_params import StorageCreateParams as StorageCreateParams
 from .storage_list_response import StorageListResponse as StorageListResponse
 from .storage_update_params import StorageUpdateParams as StorageUpdateParams
-from .storage_create_response import StorageCreateResponse as StorageCreateResponse
 from .storage_delete_response import StorageDeleteResponse as StorageDeleteResponse

together/types/beta/clusters/cluster_storage.py

@@ -1,5 +1,7 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
+from typing_extensions import Literal
+
 from ...._models import BaseModel
 
 __all__ = ["ClusterStorage"]
@@ -8,6 +10,8 @@ __all__ = ["ClusterStorage"]
 class ClusterStorage(BaseModel):
     size_tib: int
 
+    status: Literal["available", "bound", "provisioning"]
+
     volume_id: str
 
     volume_name: str

together/types/chat_completions.py

@@ -4,4 +4,4 @@ from .chat.chat_completion import ChatCompletion
 from .chat.chat_completion_chunk import ChatCompletionChunk as ChatCompletionChunk
 
 ChatCompletionResponse = ChatCompletion
-ToolCalls = ToolChoice
+ToolCalls = ToolChoice

together/types/endpoints.py

@@ -1,4 +1,4 @@
 # Manually added to minimize breaking changes from V1
 from together.types import DedicatedEndpoint as DedicatedEndpoint
 
-ListEndpoint = DedicatedEndpoint
+ListEndpoint = DedicatedEndpoint

{together-2.0.0a15.dist-info → together-2.0.0a16.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: together
-Version: 2.0.0a15
+Version: 2.0.0a16
 Summary: The official Python library for the together API
 Project-URL: Homepage, https://github.com/togethercomputer/together-py
 Project-URL: Repository, https://github.com/togethercomputer/together-py
@@ -33,12 +33,15 @@ Requires-Dist: rich>=13.7.1
 Requires-Dist: sniffio
 Requires-Dist: tabulate>=0.9.0
 Requires-Dist: tqdm>=4.67.1
+Requires-Dist: types-pyyaml>=6.0.12.20250915
 Requires-Dist: types-tabulate>=0.9.0.20240106
 Requires-Dist: types-tqdm>=4.67.0.20250516
 Requires-Dist: typing-extensions<5,>=4.10
 Provides-Extra: aiohttp
 Requires-Dist: aiohttp; extra == 'aiohttp'
 Requires-Dist: httpx-aiohttp>=0.1.9; extra == 'aiohttp'
+Provides-Extra: cli
+Requires-Dist: pyyaml; extra == 'cli'
 Provides-Extra: pyarrow
 Requires-Dist: pyarrow-stubs>=10.0.1.7; extra == 'pyarrow'
 Requires-Dist: pyarrow>=16.1.0; extra == 'pyarrow'
@@ -236,7 +239,6 @@ cluster = client.beta.clusters.create(
     billing_type="RESERVED",
     cluster_name="cluster_name",
     driver_version="CUDA_12_5_555",
-    duration_days=0,
     gpu_type="H100_SXM",
     num_gpus=0,
     region="us-central-8",