c63a5cfe-b235-4fbe-8bbb-82a9e02a482a-python 0.1.0a8__tar.gz → 0.1.0a10__tar.gz

c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10/.release-please-manifest.json

@@ -0,0 +1,3 @@
+{
+  ".": "0.1.0-alpha.10"
+}

{c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a8 → c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10}/CHANGELOG.md

@@ -1,5 +1,23 @@
 # Changelog
 
+## 0.1.0-alpha.10 (2025-06-28)
+
+Full Changelog: [v0.1.0-alpha.9...v0.1.0-alpha.10](https://github.com/digitalocean/gradientai-python/compare/v0.1.0-alpha.9...v0.1.0-alpha.10)
+
+### Features
+
+* **api:** manual updates ([0e5effc](https://github.com/digitalocean/gradientai-python/commit/0e5effc727cebe88ea38f0ec4c3fcb45ffeb4924))
+* **api:** manual updates ([d510ae0](https://github.com/digitalocean/gradientai-python/commit/d510ae03f13669af7f47093af06a00609e9b7c07))
+* **api:** manual updates ([c5bc3ca](https://github.com/digitalocean/gradientai-python/commit/c5bc3caa477945dc19bbf90661ffeea86370189d))
+
+## 0.1.0-alpha.9 (2025-06-28)
+
+Full Changelog: [v0.1.0-alpha.8...v0.1.0-alpha.9](https://github.com/digitalocean/gradientai-python/compare/v0.1.0-alpha.8...v0.1.0-alpha.9)
+
+### Features
+
+* **api:** manual updates ([e0c210a](https://github.com/digitalocean/gradientai-python/commit/e0c210a0ffde24bd2c5877689f8ab222288cc597))
+
 ## 0.1.0-alpha.8 (2025-06-27)
 
 Full Changelog: [v0.1.0-alpha.7...v0.1.0-alpha.8](https://github.com/digitalocean/gradientai-python/compare/v0.1.0-alpha.7...v0.1.0-alpha.8)

{c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a8 → c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: c63a5cfe-b235-4fbe-8bbb-82a9e02a482a-python
-Version: 0.1.0a8
+Version: 0.1.0a10
 Summary: The official Python library for GradientAI
 Project-URL: Homepage, https://github.com/digitalocean/gradientai-python
 Project-URL: Repository, https://github.com/digitalocean/gradientai-python
@@ -61,26 +61,33 @@ The full API of this library can be found in [api.md](https://github.com/digital
 import os
 from gradientai import GradientAI
 
-client = GradientAI(
+api_client = GradientAI(
     api_key=os.environ.get("GRADIENTAI_API_KEY"),  # This is the default and can be omitted
 )
+inference_client = GradientAI(
+    inference_key=os.environ.get("GRADIENTAI_INFERENCE_KEY"),  # This is the default and can be omitted
+)
 
-completion = client.agents.chat.completions.create(
+print(api_client.agents.list())
+
+completion = inference_client.chat.completions.create(
     messages=[
         {
-            "content": "string",
-            "role": "system",
+            "role": "user",
+            "content": "What is the capital of France?",
         }
     ],
-    model="llama3-8b-instruct",
+    model="llama3.3-70b-instruct",
 )
-print(completion.id)
+
+print(completion.choices[0].message)
+
 ```
 
-While you can provide an `api_key` keyword argument,
+While you can provide an `api_key`, `inference_key` keyword argument,
 we recommend using [python-dotenv](https://pypi.org/project/python-dotenv/)
-to add `GRADIENTAI_API_KEY="My API Key"` to your `.env` file
-so that your API Key is not stored in source control.
+to add `GRADIENTAI_API_KEY="My API Key"`, `GRADIENTAI_INFERENCE_KEY="My INFERENCE Key"` to your `.env` file
+so that your keys are not stored in source control.
 
 ## Async usage
 
@@ -100,13 +107,13 @@ async def main() -> None:
     completion = await client.agents.chat.completions.create(
         messages=[
             {
-                "content": "string",
-                "role": "system",
+                "role": "user",
+                "content": "What is the capital of France?",
             }
         ],
-        model="llama3-8b-instruct",
+        model="llama3.3-70b-instruct",
     )
-    print(completion.id)
+    print(completion.choices)
 
 
 asyncio.run(main())
@@ -142,18 +149,62 @@ async def main() -> None:
         completion = await client.agents.chat.completions.create(
             messages=[
                 {
-                    "content": "string",
-                    "role": "system",
+                    "role": "user",
+                    "content": "What is the capital of France?",
                 }
             ],
-            model="llama3-8b-instruct",
+            model="llama3.3-70b-instruct",
         )
-        print(completion.id)
+        print(completion.choices)
 
 
 asyncio.run(main())
 ```
 
+## Streaming responses
+
+We provide support for streaming responses using Server Side Events (SSE).
+
+```python
+from gradientai import GradientAI
+
+client = GradientAI()
+
+stream = client.agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
+    stream=True,
+)
+for completion in stream:
+    print(completion.choices)
+```
+
+The async client uses the exact same interface.
+
+```python
+from gradientai import AsyncGradientAI
+
+client = AsyncGradientAI()
+
+stream = await client.agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
+    stream=True,
+)
+async for completion in stream:
+    print(completion.choices)
+```
+
 ## Using types
 
 Nested request parameters are [TypedDicts](https://docs.python.org/3/library/typing.html#typing.TypedDict). Responses are [Pydantic models](https://docs.pydantic.dev) which also provide helper methods for things like:
@@ -201,8 +252,14 @@ from gradientai import GradientAI
 client = GradientAI()
 
 try:
-    client.agents.versions.list(
-        uuid="REPLACE_ME",
+    client.agents.chat.completions.create(
+        messages=[
+            {
+                "role": "user",
+                "content": "What is the capital of France?",
+            }
+        ],
+        model="llama3.3-70b-instruct",
     )
 except gradientai.APIConnectionError as e:
     print("The server could not be reached")
@@ -246,8 +303,14 @@ client = GradientAI(
 )
 
 # Or, configure per-request:
-client.with_options(max_retries=5).agents.versions.list(
-    uuid="REPLACE_ME",
+client.with_options(max_retries=5).agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
 )
 ```
 
@@ -271,8 +334,14 @@ client = GradientAI(
 )
 
 # Override per-request:
-client.with_options(timeout=5.0).agents.versions.list(
-    uuid="REPLACE_ME",
+client.with_options(timeout=5.0).agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
 )
 ```
 
@@ -314,13 +383,17 @@ The "raw" Response object can be accessed by prefixing `.with_raw_response.` to
 from gradientai import GradientAI
 
 client = GradientAI()
-response = client.agents.versions.with_raw_response.list(
-    uuid="REPLACE_ME",
+response = client.agents.chat.completions.with_raw_response.create(
+    messages=[{
+        "role": "user",
+        "content": "What is the capital of France?",
+    }],
+    model="llama3.3-70b-instruct",
 )
 print(response.headers.get('X-My-Header'))
 
-version = response.parse()  # get the object that `agents.versions.list()` would have returned
-print(version.agent_versions)
+completion = response.parse()  # get the object that `agents.chat.completions.create()` would have returned
+print(completion.choices)
 ```
 
 These methods return an [`APIResponse`](https://github.com/digitalocean/gradientai-python/tree/main/src/gradientai/_response.py) object.
@@ -334,8 +407,14 @@ The above interface eagerly reads the full response body when you make the reque
 To stream the response body, use `.with_streaming_response` instead, which requires a context manager and only reads the response body once you call `.read()`, `.text()`, `.json()`, `.iter_bytes()`, `.iter_text()`, `.iter_lines()` or `.parse()`. In the async client, these are async methods.
 
 ```python
-with client.agents.versions.with_streaming_response.list(
-    uuid="REPLACE_ME",
+with client.agents.chat.completions.with_streaming_response.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
 ) as response:
     print(response.headers.get("X-My-Header"))
 

{c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a8 → c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10}/README.md

@@ -27,26 +27,33 @@ The full API of this library can be found in [api.md](api.md).
 import os
 from gradientai import GradientAI
 
-client = GradientAI(
+api_client = GradientAI(
     api_key=os.environ.get("GRADIENTAI_API_KEY"),  # This is the default and can be omitted
 )
+inference_client = GradientAI(
+    inference_key=os.environ.get("GRADIENTAI_INFERENCE_KEY"),  # This is the default and can be omitted
+)
 
-completion = client.agents.chat.completions.create(
+print(api_client.agents.list())
+
+completion = inference_client.chat.completions.create(
     messages=[
         {
-            "content": "string",
-            "role": "system",
+            "role": "user",
+            "content": "What is the capital of France?",
         }
     ],
-    model="llama3-8b-instruct",
+    model="llama3.3-70b-instruct",
 )
-print(completion.id)
+
+print(completion.choices[0].message)
+
 ```
 
-While you can provide an `api_key` keyword argument,
+While you can provide an `api_key`, `inference_key` keyword argument,
 we recommend using [python-dotenv](https://pypi.org/project/python-dotenv/)
-to add `GRADIENTAI_API_KEY="My API Key"` to your `.env` file
-so that your API Key is not stored in source control.
+to add `GRADIENTAI_API_KEY="My API Key"`, `GRADIENTAI_INFERENCE_KEY="My INFERENCE Key"` to your `.env` file
+so that your keys are not stored in source control.
 
 ## Async usage
 
@@ -66,13 +73,13 @@ async def main() -> None:
     completion = await client.agents.chat.completions.create(
         messages=[
             {
-                "content": "string",
-                "role": "system",
+                "role": "user",
+                "content": "What is the capital of France?",
             }
         ],
-        model="llama3-8b-instruct",
+        model="llama3.3-70b-instruct",
     )
-    print(completion.id)
+    print(completion.choices)
 
 
 asyncio.run(main())
@@ -108,18 +115,62 @@ async def main() -> None:
         completion = await client.agents.chat.completions.create(
             messages=[
                 {
-                    "content": "string",
-                    "role": "system",
+                    "role": "user",
+                    "content": "What is the capital of France?",
                 }
             ],
-            model="llama3-8b-instruct",
+            model="llama3.3-70b-instruct",
         )
-        print(completion.id)
+        print(completion.choices)
 
 
 asyncio.run(main())
 ```
 
+## Streaming responses
+
+We provide support for streaming responses using Server Side Events (SSE).
+
+```python
+from gradientai import GradientAI
+
+client = GradientAI()
+
+stream = client.agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
+    stream=True,
+)
+for completion in stream:
+    print(completion.choices)
+```
+
+The async client uses the exact same interface.
+
+```python
+from gradientai import AsyncGradientAI
+
+client = AsyncGradientAI()
+
+stream = await client.agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
+    stream=True,
+)
+async for completion in stream:
+    print(completion.choices)
+```
+
 ## Using types
 
 Nested request parameters are [TypedDicts](https://docs.python.org/3/library/typing.html#typing.TypedDict). Responses are [Pydantic models](https://docs.pydantic.dev) which also provide helper methods for things like:
@@ -167,8 +218,14 @@ from gradientai import GradientAI
 client = GradientAI()
 
 try:
-    client.agents.versions.list(
-        uuid="REPLACE_ME",
+    client.agents.chat.completions.create(
+        messages=[
+            {
+                "role": "user",
+                "content": "What is the capital of France?",
+            }
+        ],
+        model="llama3.3-70b-instruct",
     )
 except gradientai.APIConnectionError as e:
     print("The server could not be reached")
@@ -212,8 +269,14 @@ client = GradientAI(
 )
 
 # Or, configure per-request:
-client.with_options(max_retries=5).agents.versions.list(
-    uuid="REPLACE_ME",
+client.with_options(max_retries=5).agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
 )
 ```
 
@@ -237,8 +300,14 @@ client = GradientAI(
 )
 
 # Override per-request:
-client.with_options(timeout=5.0).agents.versions.list(
-    uuid="REPLACE_ME",
+client.with_options(timeout=5.0).agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
 )
 ```
 
@@ -280,13 +349,17 @@ The "raw" Response object can be accessed by prefixing `.with_raw_response.` to
 from gradientai import GradientAI
 
 client = GradientAI()
-response = client.agents.versions.with_raw_response.list(
-    uuid="REPLACE_ME",
+response = client.agents.chat.completions.with_raw_response.create(
+    messages=[{
+        "role": "user",
+        "content": "What is the capital of France?",
+    }],
+    model="llama3.3-70b-instruct",
 )
 print(response.headers.get('X-My-Header'))
 
-version = response.parse()  # get the object that `agents.versions.list()` would have returned
-print(version.agent_versions)
+completion = response.parse()  # get the object that `agents.chat.completions.create()` would have returned
+print(completion.choices)
 ```
 
 These methods return an [`APIResponse`](https://github.com/digitalocean/gradientai-python/tree/main/src/gradientai/_response.py) object.
@@ -300,8 +373,14 @@ The above interface eagerly reads the full response body when you make the reque
 To stream the response body, use `.with_streaming_response` instead, which requires a context manager and only reads the response body once you call `.read()`, `.text()`, `.json()`, `.iter_bytes()`, `.iter_text()`, `.iter_lines()` or `.parse()`. In the async client, these are async methods.
 
 ```python
-with client.agents.versions.with_streaming_response.list(
-    uuid="REPLACE_ME",
+with client.agents.chat.completions.with_streaming_response.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
+    model="llama3.3-70b-instruct",
 ) as response:
     print(response.headers.get("X-My-Header"))
 

{c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a8 → c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10}/api.md

@@ -65,7 +65,7 @@ Methods:
 Types:
 
 ```python
-from gradientai.types.agents.chat import ChatCompletionChunk, CompletionCreateResponse
+from gradientai.types.agents.chat import AgentChatCompletionChunk, CompletionCreateResponse
 ```
 
 Methods:
@@ -253,6 +253,20 @@ Methods:
 - <code title="post /v2/gen-ai/agents/{parent_agent_uuid}/child_agents/{child_agent_uuid}">client.agents.routes.<a href="./src/gradientai/resources/agents/routes.py">add</a>(path_child_agent_uuid, \*, path_parent_agent_uuid, \*\*<a href="src/gradientai/types/agents/route_add_params.py">params</a>) -> <a href="./src/gradientai/types/agents/route_add_response.py">RouteAddResponse</a></code>
 - <code title="get /v2/gen-ai/agents/{uuid}/child_agents">client.agents.routes.<a href="./src/gradientai/resources/agents/routes.py">view</a>(uuid) -> <a href="./src/gradientai/types/agents/route_view_response.py">RouteViewResponse</a></code>
 
+# Chat
+
+## Completions
+
+Types:
+
+```python
+from gradientai.types.chat import ChatCompletionChunk, CompletionCreateResponse
+```
+
+Methods:
+
+- <code title="post /chat/completions">client.chat.completions.<a href="./src/gradientai/resources/chat/completions.py">create</a>(\*\*<a href="src/gradientai/types/chat/completion_create_params.py">params</a>) -> <a href="./src/gradientai/types/chat/completion_create_response.py">CompletionCreateResponse</a></code>
+
 # ModelProviders
 
 ## Anthropic
@@ -389,20 +403,6 @@ Methods:
 - <code title="get /v2/gen-ai/indexing_jobs/{indexing_job_uuid}/data_sources">client.knowledge_bases.indexing_jobs.<a href="./src/gradientai/resources/knowledge_bases/indexing_jobs.py">retrieve_data_sources</a>(indexing_job_uuid) -> <a href="./src/gradientai/types/knowledge_bases/indexing_job_retrieve_data_sources_response.py">IndexingJobRetrieveDataSourcesResponse</a></code>
 - <code title="put /v2/gen-ai/indexing_jobs/{uuid}/cancel">client.knowledge_bases.indexing_jobs.<a href="./src/gradientai/resources/knowledge_bases/indexing_jobs.py">update_cancel</a>(path_uuid, \*\*<a href="src/gradientai/types/knowledge_bases/indexing_job_update_cancel_params.py">params</a>) -> <a href="./src/gradientai/types/knowledge_bases/indexing_job_update_cancel_response.py">IndexingJobUpdateCancelResponse</a></code>
 
-# Chat
-
-## Completions
-
-Types:
-
-```python
-from gradientai.types.chat import CompletionCreateResponse
-```
-
-Methods:
-
-- <code title="post /chat/completions">client.chat.completions.<a href="./src/gradientai/resources/chat/completions.py">create</a>(\*\*<a href="src/gradientai/types/chat/completion_create_params.py">params</a>) -> <a href="./src/gradientai/types/chat/completion_create_response.py">CompletionCreateResponse</a></code>
-
 # Inference
 
 ## APIKeys

{c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a8 → c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "c63a5cfe-b235-4fbe-8bbb-82a9e02a482a-python"
-version = "0.1.0-alpha.8"
+version = "0.1.0-alpha.10"
 description = "The official Python library for GradientAI"
 dynamic = ["readme"]
 license = "Apache-2.0"

{c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a8 → c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10}/src/gradientai/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "gradientai"
-__version__ = "0.1.0-alpha.8"  # x-release-please-version
+__version__ = "0.1.0-alpha.10"  # x-release-please-version

{c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a8 → c63a5cfe_b235_4fbe_8bbb_82a9e02a482a_python-0.1.0a10}/src/gradientai/resources/agents/chat/completions.py

@@ -20,8 +20,8 @@ from ...._response import (
 from ...._streaming import Stream, AsyncStream
 from ...._base_client import make_request_options
 from ....types.agents.chat import completion_create_params
-from ....types.agents.chat.chat_completion_chunk import ChatCompletionChunk
 from ....types.agents.chat.completion_create_response import CompletionCreateResponse
+from ....types.agents.chat.agent_chat_completion_chunk import AgentChatCompletionChunk
 
 __all__ = ["CompletionsResource", "AsyncCompletionsResource"]
 
@@ -186,7 +186,7 @@ class CompletionsResource(SyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> Stream[ChatCompletionChunk]:
+    ) -> Stream[AgentChatCompletionChunk]:
         """
         Creates a model response for the given chat conversation.
 
@@ -299,7 +299,7 @@ class CompletionsResource(SyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> CompletionCreateResponse | Stream[ChatCompletionChunk]:
+    ) -> CompletionCreateResponse | Stream[AgentChatCompletionChunk]:
         """
         Creates a model response for the given chat conversation.
 
@@ -412,7 +412,7 @@ class CompletionsResource(SyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> CompletionCreateResponse | Stream[ChatCompletionChunk]:
+    ) -> CompletionCreateResponse | Stream[AgentChatCompletionChunk]:
         return self._post(
             "/chat/completions"
             if self._client._base_url_overridden
@@ -446,7 +446,7 @@ class CompletionsResource(SyncAPIResource):
             ),
             cast_to=CompletionCreateResponse,
             stream=stream or False,
-            stream_cls=Stream[ChatCompletionChunk],
+            stream_cls=Stream[AgentChatCompletionChunk],
         )
 
 
@@ -610,7 +610,7 @@ class AsyncCompletionsResource(AsyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> AsyncStream[ChatCompletionChunk]:
+    ) -> AsyncStream[AgentChatCompletionChunk]:
         """
         Creates a model response for the given chat conversation.
 
@@ -723,7 +723,7 @@ class AsyncCompletionsResource(AsyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> CompletionCreateResponse | AsyncStream[ChatCompletionChunk]:
+    ) -> CompletionCreateResponse | AsyncStream[AgentChatCompletionChunk]:
         """
         Creates a model response for the given chat conversation.
 
@@ -836,7 +836,7 @@ class AsyncCompletionsResource(AsyncAPIResource):
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> CompletionCreateResponse | AsyncStream[ChatCompletionChunk]:
+    ) -> CompletionCreateResponse | AsyncStream[AgentChatCompletionChunk]:
         return await self._post(
             "/chat/completions"
             if self._client._base_url_overridden
@@ -870,7 +870,7 @@ class AsyncCompletionsResource(AsyncAPIResource):
             ),
             cast_to=CompletionCreateResponse,
             stream=stream or False,
-            stream_cls=AsyncStream[ChatCompletionChunk],
+            stream_cls=AsyncStream[AgentChatCompletionChunk],
         )