release: 0.1.0-alpha.10 (#16)

Author: stainless-app[bot]

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "0.1.0-alpha.9"
+  ".": "0.1.0-alpha.10"
 }

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 76
 openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/digitalocean%2Fgradientai-e8b3cbc80e18e4f7f277010349f25e1319156704f359911dc464cc21a0d077a6.yml
 openapi_spec_hash: c773d792724f5647ae25a5ae4ccec208
-config_hash: 0bc3af28d4abd9be8bcc81f615bc832d
+config_hash: 9b44ce3fd39c43f2001bc11934e6b1b0

CHANGELOG.md

@@ -1,5 +1,15 @@
 # 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)

README.md

@@ -39,12 +39,13 @@ 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.choices[0].message)
 
 ```
@@ -72,13 +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())
@@ -114,41 +115,61 @@ 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
-Support for streaming responses are available by Server Side Events (SSE) for Serverless Inference and Agents.
-```
-import os
+## Streaming responses
+
+We provide support for streaming responses using Server Side Events (SSE).
+
+```python
 from gradientai import GradientAI
 
-client = GradientAI(
-    inference_key=os.environ.get("GRADIENTAI_INFERENCE_KEY")
-)
+client = GradientAI()
 
-response = client.chat.completions.create(
+stream = client.agents.chat.completions.create(
+    messages=[
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ],
     model="llama3.3-70b-instruct",
-    messages=[{ "role": "user", "content": "Write a story about a brave squirrel."}],
     stream=True,
 )
+for completion in stream:
+    print(completion.choices)
+```
 
-for chunk in response:
-    if len(chunk.choices) > 0:
-          if chunk.choices[0].delta.content:
-              print(chunk.choices[0].delta.content, end="", flush=True)
+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
 
@@ -197,8 +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")
@@ -242,8 +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",
 )
 ```
 
@@ -267,8 +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",
 )
 ```
 
@@ -310,13 +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.
@@ -330,8 +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"))
 

api.md

@@ -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 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>
-
 # Inference
 
 ## APIKeys

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "c63a5cfe-b235-4fbe-8bbb-82a9e02a482a-python"
-version = "0.1.0-alpha.9"
+version = "0.1.0-alpha.10"
 description = "The official Python library for GradientAI"
 dynamic = ["readme"]
 license = "Apache-2.0"