parallel-web 0.1.2__tar.gz → 0.2.0__tar.gz

{parallel_web-0.1.2 → parallel_web-0.2.0}/.gitignore

@@ -1,5 +1,4 @@
 .prism.log
-.vscode
 _dev
 
 __pycache__

parallel_web-0.2.0/.release-please-manifest.json

@@ -0,0 +1,3 @@
+{
+  ".": "0.2.0"
+}

{parallel_web-0.1.2 → parallel_web-0.2.0}/CHANGELOG.md

@@ -1,5 +1,59 @@
 # Changelog
 
+## 0.2.0 (2025-09-01)
+
+Full Changelog: [v0.1.3...v0.2.0](https://github.com/parallel-web/parallel-sdk-python/compare/v0.1.3...v0.2.0)
+
+### Features
+
+* **api:** update via SDK Studio ([b048bd7](https://github.com/parallel-web/parallel-sdk-python/commit/b048bd7e1c5a992ae274aa4b6df16a9d5b0f843e))
+* **api:** update via SDK Studio ([b9abf3c](https://github.com/parallel-web/parallel-sdk-python/commit/b9abf3c8b0e22b260149f01b1ef608924eefe735))
+* **api:** update via SDK Studio ([4326698](https://github.com/parallel-web/parallel-sdk-python/commit/43266988c2123fa1aff00bf0b62c355b0c2bf04e))
+* clean up environment call outs ([3a102e9](https://github.com/parallel-web/parallel-sdk-python/commit/3a102e9a05476e4d28c0ac386cd156cc0fe8b5cf))
+* **client:** add support for aiohttp ([4e2aa32](https://github.com/parallel-web/parallel-sdk-python/commit/4e2aa32ad8242745f56e5a8b810d33c362967dad))
+* **client:** support file upload requests ([ec0c2cf](https://github.com/parallel-web/parallel-sdk-python/commit/ec0c2cf30bd24524567232ad0f661facda124203))
+
+
+### Bug Fixes
+
+* add types for backwards compatibility ([c975302](https://github.com/parallel-web/parallel-sdk-python/commit/c975302c0d61d1d6731ccaeb7977c2009cb0b666))
+* avoid newer type syntax ([2ea196d](https://github.com/parallel-web/parallel-sdk-python/commit/2ea196d5d4c7881e61dc848a1387770b4e27e304))
+* **ci:** correct conditional ([99d37f6](https://github.com/parallel-web/parallel-sdk-python/commit/99d37f657a249987ccae60dd0e62f296ab0c1d85))
+* **ci:** release-doctor — report correct token name ([310076b](https://github.com/parallel-web/parallel-sdk-python/commit/310076b2f8a75ed29ba2a1fae0f6e840ec43bb5b))
+* **client:** don't send Content-Type header on GET requests ([f103b4a](https://github.com/parallel-web/parallel-sdk-python/commit/f103b4a72fc25f6a8dd1bda0c8d040aba1f527d1))
+* **parsing:** correctly handle nested discriminated unions ([c9a2300](https://github.com/parallel-web/parallel-sdk-python/commit/c9a23002be2d78a11b5c1b7c901f4ddb32663393))
+* **parsing:** ignore empty metadata ([ab434aa](https://github.com/parallel-web/parallel-sdk-python/commit/ab434aa7bd088fc16279255ae36138ab6dff0730))
+* **parsing:** parse extra field types ([85f5cd4](https://github.com/parallel-web/parallel-sdk-python/commit/85f5cd4191ae168ed443e78a2c7bd747d51404b3))
+
+
+### Chores
+
+* **ci:** change upload type ([40dbd3b](https://github.com/parallel-web/parallel-sdk-python/commit/40dbd3b7d5becf0fe54b62a4acd8696957380053))
+* **ci:** only run for pushes and fork pull requests ([d55fbea](https://github.com/parallel-web/parallel-sdk-python/commit/d55fbea54037d2d833ecc281cbddbc8d6700d24d))
+* **internal:** add Sequence related utils ([cb9a7a9](https://github.com/parallel-web/parallel-sdk-python/commit/cb9a7a905ca4a4a9ba35e540f6c47a8bf89c87d2))
+* **internal:** bump pinned h11 dep ([818f1dd](https://github.com/parallel-web/parallel-sdk-python/commit/818f1ddb3ba1be6bfdb9aee1322d6a3d8a98667a))
+* **internal:** change ci workflow machines ([a90da34](https://github.com/parallel-web/parallel-sdk-python/commit/a90da34910585453eac918a5f273749c00d2f743))
+* **internal:** codegen related update ([47ea68b](https://github.com/parallel-web/parallel-sdk-python/commit/47ea68bd44ad52ac1c18e7215c013f408914890c))
+* **internal:** fix ruff target version ([4e5dbda](https://github.com/parallel-web/parallel-sdk-python/commit/4e5dbda03907f45ac31d18d89714e86f26e79866))
+* **internal:** update comment in script ([631b045](https://github.com/parallel-web/parallel-sdk-python/commit/631b045ae2f138e4c8098fafd9466451d61ca82a))
+* **internal:** update pyright exclude list ([8d2fb29](https://github.com/parallel-web/parallel-sdk-python/commit/8d2fb29b5d80a2fa9ee81a6f9510134fb7bab908))
+* **internal:** version bump ([90d26a5](https://github.com/parallel-web/parallel-sdk-python/commit/90d26a5e8db8bd6a27f9bbc96595da87bd7ea0f3))
+* **package:** mark python 3.13 as supported ([6fa54c4](https://github.com/parallel-web/parallel-sdk-python/commit/6fa54c42a17f5e731f5e97214f0212a0828d3cb8))
+* **project:** add settings file for vscode ([acdeda2](https://github.com/parallel-web/parallel-sdk-python/commit/acdeda2f1f95f5bade2da52d5a2aa8560e71369d))
+* **readme:** fix version rendering on pypi ([2bf10b0](https://github.com/parallel-web/parallel-sdk-python/commit/2bf10b073ab7e015b08c106d265a9091752df51a))
+* **readme:** Remove references to methods, update FAQ for beta ([cefefbf](https://github.com/parallel-web/parallel-sdk-python/commit/cefefbfccba78fdabcc925728836d70400d4e5aa))
+* **tests:** skip some failing tests on the latest python versions ([13b1533](https://github.com/parallel-web/parallel-sdk-python/commit/13b153381e9b7c998a7ebef878518222678dfa83))
+* update @stainless-api/prism-cli to v5.15.0 ([56b5aab](https://github.com/parallel-web/parallel-sdk-python/commit/56b5aab87a833c27b8e1a2bc7c4bf2169ee281a8))
+* update github action ([3d90e19](https://github.com/parallel-web/parallel-sdk-python/commit/3d90e196184e540242fb310cc55b0219d20dff45))
+
+## 0.1.3 (2025-08-09)
+
+Full Changelog: [v0.1.2...v0.1.3](https://github.com/parallel-web/parallel-sdk-python/compare/v0.1.2...v0.1.3)
+
+### Chores
+
+* **readme:** update descriptions ([3212a0f](https://github.com/parallel-web/parallel-sdk-python/commit/3212a0fc32d744e7df3d0dcedf527b176a73a91b))
+
 ## 0.1.2 (2025-06-25)
 
 Full Changelog: [v0.1.1...v0.1.2](https://github.com/parallel-web/parallel-sdk-python/compare/v0.1.1...v0.1.2)

{parallel_web-0.1.2 → parallel_web-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: parallel-web
-Version: 0.1.2
+Version: 0.2.0
 Summary: The official Python library for the Parallel API
 Project-URL: Homepage, https://github.com/parallel-web/parallel-sdk-python
 Project-URL: Repository, https://github.com/parallel-web/parallel-sdk-python
@@ -18,6 +18,7 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Classifier: Typing :: Typed
 Requires-Python: >=3.8
@@ -27,11 +28,15 @@ Requires-Dist: httpx<1,>=0.23.0
 Requires-Dist: pydantic<3,>=1.9.0
 Requires-Dist: sniffio
 Requires-Dist: typing-extensions<5,>=4.10
+Provides-Extra: aiohttp
+Requires-Dist: aiohttp; extra == 'aiohttp'
+Requires-Dist: httpx-aiohttp>=0.1.8; extra == 'aiohttp'
 Description-Content-Type: text/markdown
 
 # Parallel Python API library
 
-[![PyPI version](https://github.com/parallel-web/parallel-sdk-python/tree/main/<https://img.shields.io/pypi/v/parallel-web.svg?label=pypi%20(stable)>)](https://pypi.org/project/parallel-web/)
+<!-- prettier-ignore -->
+[![PyPI version](https://img.shields.io/pypi/v/parallel-web.svg?label=pypi%20(stable))](https://pypi.org/project/parallel-web/)
 
 The Parallel Python library provides convenient access to the Parallel REST API from any Python 3.8+
 application. The library includes type definitions for all request params and response fields,
@@ -42,7 +47,7 @@ It is generated with [Stainless](https://www.stainless.com/).
 
 ## Documentation
 
-The REST API documentation can be found on our [docs](https://docs.parallel.ai).
+The REST API documentation can be found in our [docs](https://docs.parallel.ai).
 The full API of this Python library can be found in [api.md](https://github.com/parallel-web/parallel-sdk-python/tree/main/api.md).
 
 ## Installation
@@ -64,12 +69,12 @@ client = Parallel(
     api_key=os.environ.get("PARALLEL_API_KEY"),  # This is the default and can be omitted
 )
 
-run_result = client.task_run.execute(
+task_run = client.task_run.create(
     input="France (2023)",
     processor="core",
-    output="GDP"
 )
-print(run_result.output)
+task_run_result = client.task_run.result(run_id=task_run.run_id)
+print(task_run_result.output)
 ```
 
 While you can provide an `api_key` keyword argument,
@@ -77,6 +82,11 @@ we recommend using [python-dotenv](https://pypi.org/project/python-dotenv/)
 to add `PARALLEL_API_KEY="My API Key"` to your `.env` file
 so that your API Key is not stored in source control.
 
+The API also supports typed inputs and outputs via Pydantic objects. See the relevant
+section on [convenience methods](https://github.com/parallel-web/parallel-sdk-python/tree/main/#convenience-methods).
+
+For information on what tasks are and how to specify them, see [our docs](https://docs.parallel.ai/task-api/core-concepts/specify-a-task).
+
 ## Async usage
 
 Simply import `AsyncParallel` instead of `Parallel` and use `await` with each API call:
@@ -92,12 +102,9 @@ client = AsyncParallel(
 
 
 async def main() -> None:
-    run_result = await client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output="GDP"
-    )
-    print(run_result.output)
+    task_run = await client.task_run.create(input="France (2023)", processor="core")
+    run_result = await client.task_run.result(run_id=task_run.run_id)
+    print(run_result.output.content)
 
 
 if __name__ == "__main__":
@@ -106,115 +113,35 @@ if __name__ == "__main__":
 
 To get the best performance out of Parallel's API, we recommend
 using the asynchronous client, especially for executing multiple Task Runs concurrently.
-Functionality between the synchronous and asynchronous clients is identical.
+Functionality between the synchronous and asynchronous clients is identical, including
+the convenience methods.
 
-## Convenience methods
+## Frequently Asked Questions
 
-### Execute
+**Does the Task API accept prompts or objectives?**
 
-The `execute` method provides a single call which combines creating a task run,
-polling until it is completed, and parsing structured outputs (if specified).
+No, there are no `objective` or `prompt` parameters that can be specified for calls to
+the Task API. Instead, provide any directives or instructions via the schemas. For
+more information, check [our docs](https://docs.parallel.ai/task-api/core-concepts/specify-a-task).
 
-If an output type which inherits from `BaseModel` is
-specified in the call to `.execute()`, the response content will be parsed into an
-instance of the provided output type. The parsed output can be accessed via the
-`parsed` property on the output field of the response.
+**Can I access beta parameters or endpoints via the SDK?**
 
-```python
-import os
-import asyncio
-from parallel import AsyncParallel
-from pydantic import BaseModel
+Yes, the SDK supports both beta endpoints and beta header parameters for the Task API.
+All beta parameters are accessible via the `client.beta` namespace in the SDK.
 
-client = AsyncParallel()
+**Can I specify a timeout for API calls?**
 
-class SampleOutputStructure(BaseModel):
-    output: str
+Yes, all methods support a timeout. For more information, see [Timeouts](https://github.com/parallel-web/parallel-sdk-python/tree/main/#timeouts).
 
-async def main() -> None:
-    # with pydantic
-    run_result = await client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output=SampleOutputStructure,
-    )
-    # parsed output of type SampleOutputStructure
-    print(run_result.output.parsed)
-    # without pydantic
-    run_result = await client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output="GDP"
-    )
-    print(run_result.output)
+**Can I specify retries via the SDK?**
 
+Yes, errors can be retried via the SDK — the default retry count is 2. The maximum number
+of retries can be configured at the client level. For information on which errors
+are automatically retried and how to configure retry settings, see [Retries](https://github.com/parallel-web/parallel-sdk-python/tree/main/#retries).
 
-if __name__ == "__main__":
-    asyncio.run(main())
-```
+## Low‑level API access
 
-The async client allows creating several task runs without blocking.
-To create multiple task runs in one go, call execute and then gather the results at the end.
-
-```python
-import asyncio
-import os
-
-from parallel import AsyncParallel
-from pydantic import BaseModel, Field
-from typing import List
-
-class CountryInput(BaseModel):
-    country: str = Field(
-        description="Name of the country to research. Must be a recognized "
-        "sovereign nation (e.g., 'France', 'Japan')."
-    )
-    year: int = Field(
-        description="Year for which to retrieve data. Must be 2000 or later. "
-                    "Use most recent full-year estimates if year is current."
-    )
-
-class CountryOutput(BaseModel):
-    gdp: str = Field(
-        description="GDP in USD for the year, formatted like '$3.1 trillion (2023)'."
-    )
-    top_exports: List[str] = Field(
-        description="Top 3 exported goods/services by value. Use credible sources."
-    )
-    top_imports: List[str] = Field(
-        description="Top 3 imported goods/services by value. Use credible sources."
-    )
-
-async def main():
-    # Initialize the Parallel client
-    client = AsyncParallel(api_key=os.environ.get("PARALLEL_API_KEY"))
-
-    # Prepare structured input
-    input_data = [
-        CountryInput(country="France", year=2023),
-        CountryInput(country="Germany", year=2023),
-        CountryInput(country="Italy", year=2023)
-    ]
-
-    run_results = await asyncio.gather(*[
-        client.task_run.execute(
-            input=datum,
-            output=CountryOutput,
-            processor="core"
-        )
-        for datum in input_data
-    ])
-
-    for run_input, run_result in zip(input_data, run_results):
-        print(f"Task run output for {run_input}: {run_result.output.parsed}")
-
-if __name__ == "__main__":
-    asyncio.run(main())
-```
-
-## Low level API Access
-
-The library also provides access to the low level API for accessing the Parallel API.
+The library also provides low‑level access to the Parallel API.
 
 ```python
 from parallel import Parallel
@@ -262,13 +189,13 @@ task_run = client.task_run.create(
 )
 
 run_result = client.task_run.result(task_run.run_id)
-print(run_result.output)
+print(run_result.output.content)
 ```
 
 For more information, please check out the relevant section in our docs:
 
-- [Task Spec](https://docs.parallel.ai/core-concepts/task-spec)
-- [Task Runs](https://docs.parallel.ai/core-concepts/task-runs)
+- [Task Spec](https://docs.parallel.ai/task-api/core-concepts/specify-a-task)
+- [Task Runs](https://docs.parallel.ai/task-api/core-concepts/execute-task-run)
 
 ## Handling errors
 
@@ -286,11 +213,7 @@ from parallel import Parallel
 client = Parallel()
 
 try:
-    client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output="GDP"
-    )
+    client.task_run.create(input="France (2023)", processor="core")
 except parallel.APIConnectionError as e:
     print("The server could not be reached")
     print(e.__cause__)  # an underlying Exception, likely raised within httpx.
@@ -333,11 +256,7 @@ client = Parallel(
 )
 
 # Or, configure per-request:
-client.with_options(max_retries=5).task_run.execute(
-    input="France (2023)",
-    processor="core",
-    output="GDP"
-)
+client.with_options(max_retries=5).task_run.create(input="France (2023)", processor="core")
 ```
 
 ### Timeouts
@@ -360,11 +279,7 @@ client = Parallel(
 )
 
 # Override per-request:
-client.with_options(timeout=5.0).task_run.execute(
-    input="France (2023)",
-    processor="core",
-    output="GDP"
-)
+client.with_options(timeout=5.0).task_run.create(input="France (2023)", processor="core")
 ```
 
 On timeout, an `APITimeoutError` is thrown.
@@ -405,15 +320,14 @@ The "raw" Response object can be accessed by prefixing `.with_raw_response.` to
 from parallel import Parallel
 
 client = Parallel()
-response = client.task_run.with_raw_response.execute(
+response = client.task_run.with_raw_response.create(
     input="France (2023)",
     processor="core",
-    output="GDP"
 )
 print(response.headers.get('X-My-Header'))
 
-task_run = response.parse()  # get the object that `task_run.execute()` would have returned
-print(task_run.output)
+task_run = response.parse()
+print(task_run.run_id)
 ```
 
 These methods return an [`APIResponse`](https://github.com/parallel-web/parallel-sdk-python/tree/main/src/parallel/_response.py) object.
@@ -427,10 +341,8 @@ 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.task_run.with_streaming_response.execute(
-    input="France (2023)",
-    processor="core",
-    output="GDP"
+with client.task_run.with_streaming_response.create(
+    input="France (2023)", processor="core"
 ) as response:
     print(response.headers.get("X-My-Header"))
 

{parallel_web-0.1.2 → parallel_web-0.2.0}/README.md

@@ -1,6 +1,7 @@
 # Parallel Python API library
 
-[![PyPI version](<https://img.shields.io/pypi/v/parallel-web.svg?label=pypi%20(stable)>)](https://pypi.org/project/parallel-web/)
+<!-- prettier-ignore -->
+[![PyPI version](https://img.shields.io/pypi/v/parallel-web.svg?label=pypi%20(stable))](https://pypi.org/project/parallel-web/)
 
 The Parallel Python library provides convenient access to the Parallel REST API from any Python 3.8+
 application. The library includes type definitions for all request params and response fields,
@@ -11,7 +12,7 @@ It is generated with [Stainless](https://www.stainless.com/).
 
 ## Documentation
 
-The REST API documentation can be found on our [docs](https://docs.parallel.ai).
+The REST API documentation can be found in our [docs](https://docs.parallel.ai).
 The full API of this Python library can be found in [api.md](api.md).
 
 ## Installation
@@ -33,12 +34,12 @@ client = Parallel(
     api_key=os.environ.get("PARALLEL_API_KEY"),  # This is the default and can be omitted
 )
 
-run_result = client.task_run.execute(
+task_run = client.task_run.create(
     input="France (2023)",
     processor="core",
-    output="GDP"
 )
-print(run_result.output)
+task_run_result = client.task_run.result(run_id=task_run.run_id)
+print(task_run_result.output)
 ```
 
 While you can provide an `api_key` keyword argument,
@@ -46,6 +47,11 @@ we recommend using [python-dotenv](https://pypi.org/project/python-dotenv/)
 to add `PARALLEL_API_KEY="My API Key"` to your `.env` file
 so that your API Key is not stored in source control.
 
+The API also supports typed inputs and outputs via Pydantic objects. See the relevant
+section on [convenience methods](#convenience-methods).
+
+For information on what tasks are and how to specify them, see [our docs](https://docs.parallel.ai/task-api/core-concepts/specify-a-task).
+
 ## Async usage
 
 Simply import `AsyncParallel` instead of `Parallel` and use `await` with each API call:
@@ -61,12 +67,9 @@ client = AsyncParallel(
 
 
 async def main() -> None:
-    run_result = await client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output="GDP"
-    )
-    print(run_result.output)
+    task_run = await client.task_run.create(input="France (2023)", processor="core")
+    run_result = await client.task_run.result(run_id=task_run.run_id)
+    print(run_result.output.content)
 
 
 if __name__ == "__main__":
@@ -75,115 +78,35 @@ if __name__ == "__main__":
 
 To get the best performance out of Parallel's API, we recommend
 using the asynchronous client, especially for executing multiple Task Runs concurrently.
-Functionality between the synchronous and asynchronous clients is identical.
+Functionality between the synchronous and asynchronous clients is identical, including
+the convenience methods.
 
-## Convenience methods
+## Frequently Asked Questions
 
-### Execute
+**Does the Task API accept prompts or objectives?**
 
-The `execute` method provides a single call which combines creating a task run,
-polling until it is completed, and parsing structured outputs (if specified).
+No, there are no `objective` or `prompt` parameters that can be specified for calls to
+the Task API. Instead, provide any directives or instructions via the schemas. For
+more information, check [our docs](https://docs.parallel.ai/task-api/core-concepts/specify-a-task).
 
-If an output type which inherits from `BaseModel` is
-specified in the call to `.execute()`, the response content will be parsed into an
-instance of the provided output type. The parsed output can be accessed via the
-`parsed` property on the output field of the response.
+**Can I access beta parameters or endpoints via the SDK?**
 
-```python
-import os
-import asyncio
-from parallel import AsyncParallel
-from pydantic import BaseModel
+Yes, the SDK supports both beta endpoints and beta header parameters for the Task API.
+All beta parameters are accessible via the `client.beta` namespace in the SDK.
 
-client = AsyncParallel()
+**Can I specify a timeout for API calls?**
 
-class SampleOutputStructure(BaseModel):
-    output: str
+Yes, all methods support a timeout. For more information, see [Timeouts](#timeouts).
 
-async def main() -> None:
-    # with pydantic
-    run_result = await client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output=SampleOutputStructure,
-    )
-    # parsed output of type SampleOutputStructure
-    print(run_result.output.parsed)
-    # without pydantic
-    run_result = await client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output="GDP"
-    )
-    print(run_result.output)
+**Can I specify retries via the SDK?**
 
+Yes, errors can be retried via the SDK — the default retry count is 2. The maximum number
+of retries can be configured at the client level. For information on which errors
+are automatically retried and how to configure retry settings, see [Retries](#retries).
 
-if __name__ == "__main__":
-    asyncio.run(main())
-```
+## Low‑level API access
 
-The async client allows creating several task runs without blocking.
-To create multiple task runs in one go, call execute and then gather the results at the end.
-
-```python
-import asyncio
-import os
-
-from parallel import AsyncParallel
-from pydantic import BaseModel, Field
-from typing import List
-
-class CountryInput(BaseModel):
-    country: str = Field(
-        description="Name of the country to research. Must be a recognized "
-        "sovereign nation (e.g., 'France', 'Japan')."
-    )
-    year: int = Field(
-        description="Year for which to retrieve data. Must be 2000 or later. "
-                    "Use most recent full-year estimates if year is current."
-    )
-
-class CountryOutput(BaseModel):
-    gdp: str = Field(
-        description="GDP in USD for the year, formatted like '$3.1 trillion (2023)'."
-    )
-    top_exports: List[str] = Field(
-        description="Top 3 exported goods/services by value. Use credible sources."
-    )
-    top_imports: List[str] = Field(
-        description="Top 3 imported goods/services by value. Use credible sources."
-    )
-
-async def main():
-    # Initialize the Parallel client
-    client = AsyncParallel(api_key=os.environ.get("PARALLEL_API_KEY"))
-
-    # Prepare structured input
-    input_data = [
-        CountryInput(country="France", year=2023),
-        CountryInput(country="Germany", year=2023),
-        CountryInput(country="Italy", year=2023)
-    ]
-
-    run_results = await asyncio.gather(*[
-        client.task_run.execute(
-            input=datum,
-            output=CountryOutput,
-            processor="core"
-        )
-        for datum in input_data
-    ])
-
-    for run_input, run_result in zip(input_data, run_results):
-        print(f"Task run output for {run_input}: {run_result.output.parsed}")
-
-if __name__ == "__main__":
-    asyncio.run(main())
-```
-
-## Low level API Access
-
-The library also provides access to the low level API for accessing the Parallel API.
+The library also provides low‑level access to the Parallel API.
 
 ```python
 from parallel import Parallel
@@ -231,13 +154,13 @@ task_run = client.task_run.create(
 )
 
 run_result = client.task_run.result(task_run.run_id)
-print(run_result.output)
+print(run_result.output.content)
 ```
 
 For more information, please check out the relevant section in our docs:
 
-- [Task Spec](https://docs.parallel.ai/core-concepts/task-spec)
-- [Task Runs](https://docs.parallel.ai/core-concepts/task-runs)
+- [Task Spec](https://docs.parallel.ai/task-api/core-concepts/specify-a-task)
+- [Task Runs](https://docs.parallel.ai/task-api/core-concepts/execute-task-run)
 
 ## Handling errors
 
@@ -255,11 +178,7 @@ from parallel import Parallel
 client = Parallel()
 
 try:
-    client.task_run.execute(
-        input="France (2023)",
-        processor="core",
-        output="GDP"
-    )
+    client.task_run.create(input="France (2023)", processor="core")
 except parallel.APIConnectionError as e:
     print("The server could not be reached")
     print(e.__cause__)  # an underlying Exception, likely raised within httpx.
@@ -302,11 +221,7 @@ client = Parallel(
 )
 
 # Or, configure per-request:
-client.with_options(max_retries=5).task_run.execute(
-    input="France (2023)",
-    processor="core",
-    output="GDP"
-)
+client.with_options(max_retries=5).task_run.create(input="France (2023)", processor="core")
 ```
 
 ### Timeouts
@@ -329,11 +244,7 @@ client = Parallel(
 )
 
 # Override per-request:
-client.with_options(timeout=5.0).task_run.execute(
-    input="France (2023)",
-    processor="core",
-    output="GDP"
-)
+client.with_options(timeout=5.0).task_run.create(input="France (2023)", processor="core")
 ```
 
 On timeout, an `APITimeoutError` is thrown.
@@ -374,15 +285,14 @@ The "raw" Response object can be accessed by prefixing `.with_raw_response.` to
 from parallel import Parallel
 
 client = Parallel()
-response = client.task_run.with_raw_response.execute(
+response = client.task_run.with_raw_response.create(
     input="France (2023)",
     processor="core",
-    output="GDP"
 )
 print(response.headers.get('X-My-Header'))
 
-task_run = response.parse()  # get the object that `task_run.execute()` would have returned
-print(task_run.output)
+task_run = response.parse()
+print(task_run.run_id)
 ```
 
 These methods return an [`APIResponse`](https://github.com/parallel-web/parallel-sdk-python/tree/main/src/parallel/_response.py) object.
@@ -396,10 +306,8 @@ 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.task_run.with_streaming_response.execute(
-    input="France (2023)",
-    processor="core",
-    output="GDP"
+with client.task_run.with_streaming_response.create(
+    input="France (2023)", processor="core"
 ) as response:
     print(response.headers.get("X-My-Header"))