release: 2.7.0 (#320)

* codegen metadata

* feat(api): add generate_audio, image_reference, media params to videos.create

* release: 2.7.0

---------

Co-authored-by: stainless-app[bot] <142633134+stainless-app[bot]@users.noreply.github.com>

Author: stainless-app[bot]

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "2.6.0"
+  ".": "2.7.0"
 }

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 74
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/togetherai%2Ftogetherai-d91fd4a37aa010c545f29c4413b9aef8f556d0a63d3c0e5a7ee36abf1671142d.yml
-openapi_spec_hash: d3365b580d695d006648441db545aff8
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/togetherai%2Ftogetherai-76f8801298719cc87e9dc4c64b321bcfd432416d76488499d340b4bb6bf81b9b.yml
+openapi_spec_hash: ce0b83ef0a5f174461bd7d13a379b636
 config_hash: 52d213100a0ca1a4b2cdcd2718936b51

CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 2.7.0 (2026-04-03)
+
+Full Changelog: [v2.6.0...v2.7.0](https://github.com/togethercomputer/together-py/compare/v2.6.0...v2.7.0)
+
+### Features
+
+* **api:** add generate_audio, image_reference, media params to videos.create ([b2c5ca7](https://github.com/togethercomputer/together-py/commit/b2c5ca776fa2520e99b4e7c6de8bd1421b4de33f))
+
 ## 2.6.0 (2026-03-31)
 
 Full Changelog: [v2.5.0...v2.6.0](https://github.com/togethercomputer/together-py/compare/v2.5.0...v2.6.0)

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "together"
-version = "2.6.0"
+version = "2.7.0"
 description = "The official Python library for the together API"
 dynamic = ["readme"]
 license = "Apache-2.0"

src/together/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "together"
-__version__ = "2.6.0"  # x-release-please-version
+__version__ = "2.7.0"  # x-release-please-version

src/together/resources/videos.py

@@ -50,13 +50,17 @@ def create(
         model: str,
         fps: int | Omit = omit,
         frame_images: Iterable[video_create_params.FrameImage] | Omit = omit,
+        generate_audio: bool | Omit = omit,
         guidance_scale: int | Omit = omit,
         height: int | Omit = omit,
+        media: video_create_params.Media | Omit = omit,
         negative_prompt: str | Omit = omit,
         output_format: Literal["MP4", "WEBM"] | Omit = omit,
         output_quality: int | Omit = omit,
         prompt: str | Omit = omit,
+        ratio: str | Omit = omit,
         reference_images: SequenceNotStr[str] | Omit = omit,
+        resolution: str | Omit = omit,
         seconds: str | Omit = omit,
         seed: int | Omit = omit,
         steps: int | Omit = omit,
@@ -76,14 +80,20 @@ def create(
 
           fps: Frames per second. Defaults to 24.
 
-          frame_images: Array of images to guide video generation, similar to keyframes.
+          frame_images: Deprecated: use media.frame_images instead. Array of images to guide video
+              generation, similar to keyframes.
+
+          generate_audio: Whether to generate audio for the video.
 
           guidance_scale: Controls how closely the video generation follows your prompt. Higher values
               make the model adhere more strictly to your text description, while lower values
               allow more creative freedom. guidence_scale affects both visual content and
               temporal consistency.Recommended range is 6.0-10.0 for most video models. Values
               above 12 may cause over-guidance artifacts or unnatural motion patterns.
 
+          media: Media inputs for video generation. The accepted fields depend on the model type
+              (e.g. i2v, r2v, t2v, videoedit).
+
           negative_prompt: Similar to prompt, but specifies what to avoid instead of what to include
 
           output_format: Specifies the format of the output video. Defaults to MP4.
@@ -92,9 +102,13 @@ def create(
 
           prompt: Text prompt that describes the video to generate.
 
-          reference_images: Unlike frame_images which constrain specific timeline positions, reference
-              images guide the general appearance that should appear consistently across the
-              video.
+          ratio: Aspect ratio of the video.
+
+          reference_images: Deprecated: use media.reference_images instead. Unlike frame_images which
+              constrain specific timeline positions, reference images guide the general
+              appearance that should appear consistently across the video.
+
+          resolution: Video resolution.
 
           seconds: Clip duration in seconds.
 
@@ -121,13 +135,17 @@ def create(
                     "model": model,
                     "fps": fps,
                     "frame_images": frame_images,
+                    "generate_audio": generate_audio,
                     "guidance_scale": guidance_scale,
                     "height": height,
+                    "media": media,
                     "negative_prompt": negative_prompt,
                     "output_format": output_format,
                     "output_quality": output_quality,
                     "prompt": prompt,
+                    "ratio": ratio,
                     "reference_images": reference_images,
+                    "resolution": resolution,
                     "seconds": seconds,
                     "seed": seed,
                     "steps": steps,
@@ -204,13 +222,17 @@ async def create(
         model: str,
         fps: int | Omit = omit,
         frame_images: Iterable[video_create_params.FrameImage] | Omit = omit,
+        generate_audio: bool | Omit = omit,
         guidance_scale: int | Omit = omit,
         height: int | Omit = omit,
+        media: video_create_params.Media | Omit = omit,
         negative_prompt: str | Omit = omit,
         output_format: Literal["MP4", "WEBM"] | Omit = omit,
         output_quality: int | Omit = omit,
         prompt: str | Omit = omit,
+        ratio: str | Omit = omit,
         reference_images: SequenceNotStr[str] | Omit = omit,
+        resolution: str | Omit = omit,
         seconds: str | Omit = omit,
         seed: int | Omit = omit,
         steps: int | Omit = omit,
@@ -230,14 +252,20 @@ async def create(
 
           fps: Frames per second. Defaults to 24.
 
-          frame_images: Array of images to guide video generation, similar to keyframes.
+          frame_images: Deprecated: use media.frame_images instead. Array of images to guide video
+              generation, similar to keyframes.
+
+          generate_audio: Whether to generate audio for the video.
 
           guidance_scale: Controls how closely the video generation follows your prompt. Higher values
               make the model adhere more strictly to your text description, while lower values
               allow more creative freedom. guidence_scale affects both visual content and
               temporal consistency.Recommended range is 6.0-10.0 for most video models. Values
               above 12 may cause over-guidance artifacts or unnatural motion patterns.
 
+          media: Media inputs for video generation. The accepted fields depend on the model type
+              (e.g. i2v, r2v, t2v, videoedit).
+
           negative_prompt: Similar to prompt, but specifies what to avoid instead of what to include
 
           output_format: Specifies the format of the output video. Defaults to MP4.
@@ -246,9 +274,13 @@ async def create(
 
           prompt: Text prompt that describes the video to generate.
 
-          reference_images: Unlike frame_images which constrain specific timeline positions, reference
-              images guide the general appearance that should appear consistently across the
-              video.
+          ratio: Aspect ratio of the video.
+
+          reference_images: Deprecated: use media.reference_images instead. Unlike frame_images which
+              constrain specific timeline positions, reference images guide the general
+              appearance that should appear consistently across the video.
+
+          resolution: Video resolution.
 
           seconds: Clip duration in seconds.
 
@@ -275,13 +307,17 @@ async def create(
                     "model": model,
                     "fps": fps,
                     "frame_images": frame_images,
+                    "generate_audio": generate_audio,
                     "guidance_scale": guidance_scale,
                     "height": height,
+                    "media": media,
                     "negative_prompt": negative_prompt,
                     "output_format": output_format,
                     "output_quality": output_quality,
                     "prompt": prompt,
+                    "ratio": ratio,
                     "reference_images": reference_images,
+                    "resolution": resolution,
                     "seconds": seconds,
                     "seed": seed,
                     "steps": steps,

src/together/types/video_create_params.py

@@ -7,7 +7,16 @@
 
 from .._types import SequenceNotStr
 
-__all__ = ["VideoCreateParams", "FrameImage"]
+__all__ = [
+    "VideoCreateParams",
+    "FrameImage",
+    "Media",
+    "MediaAudioInput",
+    "MediaFrameImage",
+    "MediaFrameVideo",
+    "MediaReferenceVideo",
+    "MediaSourceVideo",
+]
 
 
 class VideoCreateParams(TypedDict, total=False):
@@ -18,7 +27,13 @@ class VideoCreateParams(TypedDict, total=False):
     """Frames per second. Defaults to 24."""
 
     frame_images: Iterable[FrameImage]
-    """Array of images to guide video generation, similar to keyframes."""
+    """Deprecated: use media.frame_images instead.
+
+    Array of images to guide video generation, similar to keyframes.
+    """
+
+    generate_audio: bool
+    """Whether to generate audio for the video."""
 
     guidance_scale: int
     """Controls how closely the video generation follows your prompt.
@@ -32,6 +47,12 @@ class VideoCreateParams(TypedDict, total=False):
 
     height: int
 
+    media: Media
+    """Media inputs for video generation.
+
+    The accepted fields depend on the model type (e.g. i2v, r2v, t2v, videoedit).
+    """
+
     negative_prompt: str
     """Similar to prompt, but specifies what to avoid instead of what to include"""
 
@@ -44,13 +65,20 @@ class VideoCreateParams(TypedDict, total=False):
     prompt: str
     """Text prompt that describes the video to generate."""
 
+    ratio: str
+    """Aspect ratio of the video."""
+
     reference_images: SequenceNotStr[str]
-    """
+    """Deprecated: use media.reference_images instead.
+
     Unlike frame_images which constrain specific timeline positions, reference
     images guide the general appearance that should appear consistently across the
     video.
     """
 
+    resolution: str
+    """Video resolution."""
+
     seconds: str
     """Clip duration in seconds."""
 
@@ -84,3 +112,65 @@ class FrameImage(TypedDict, total=False):
     - If size is two, frames are first and last.
     - If size is larger, frames are first, last and evenly spaced between.
     """
+
+
+class MediaAudioInput(TypedDict, total=False):
+    audio: Required[str]
+    """URL of the audio."""
+
+
+class MediaFrameImage(TypedDict, total=False):
+    input_image: Required[str]
+    """URL path to hosted image that is used for a frame"""
+
+    frame: Union[float, Literal["first", "last"]]
+    """Optional param to specify where to insert the frame.
+
+    If this is omitted, the following heuristics are applied:
+
+    - frame_images size is one, frame is first.
+    - If size is two, frames are first and last.
+    - If size is larger, frames are first, last and evenly spaced between.
+    """
+
+
+class MediaFrameVideo(TypedDict, total=False):
+    video: Required[str]
+    """URL of the video."""
+
+
+class MediaReferenceVideo(TypedDict, total=False):
+    video: Required[str]
+    """URL of the video."""
+
+
+class MediaSourceVideo(TypedDict, total=False):
+    """Source video to edit."""
+
+    video: Required[str]
+    """URL of the video."""
+
+
+class Media(TypedDict, total=False):
+    """Media inputs for video generation.
+
+    The accepted fields depend on the model type (e.g. i2v, r2v, t2v, videoedit).
+    """
+
+    audio_inputs: Iterable[MediaAudioInput]
+    """Array of audio inputs."""
+
+    frame_images: Iterable[MediaFrameImage]
+    """Array of images to guide video generation at specific timeline positions."""
+
+    frame_videos: Iterable[MediaFrameVideo]
+    """Array of video clips to use as starting clips."""
+
+    reference_images: SequenceNotStr[str]
+    """Array of image URLs that guide the general appearance across the video."""
+
+    reference_videos: Iterable[MediaReferenceVideo]
+    """Array of reference videos."""
+
+    source_video: MediaSourceVideo
+    """Source video to edit."""

tests/api_resources/test_videos.py

@@ -35,13 +35,29 @@ def test_method_create_with_all_params(self, client: Together) -> None:
                     "frame": 0,
                 }
             ],
+            generate_audio=True,
             guidance_scale=0,
             height=0,
+            media={
+                "audio_inputs": [{"audio": "audio"}],
+                "frame_images": [
+                    {
+                        "input_image": "input_image",
+                        "frame": 0,
+                    }
+                ],
+                "frame_videos": [{"video": "video"}],
+                "reference_images": ["string"],
+                "reference_videos": [{"video": "video"}],
+                "source_video": {"video": "video"},
+            },
             negative_prompt="negative_prompt",
             output_format="MP4",
             output_quality=0,
             prompt="x",
+            ratio="ratio",
             reference_images=["string"],
+            resolution="resolution",
             seconds="seconds",
             seed=0,
             steps=10,
@@ -135,13 +151,29 @@ async def test_method_create_with_all_params(self, async_client: AsyncTogether)
                     "frame": 0,
                 }
             ],
+            generate_audio=True,
             guidance_scale=0,
             height=0,
+            media={
+                "audio_inputs": [{"audio": "audio"}],
+                "frame_images": [
+                    {
+                        "input_image": "input_image",
+                        "frame": 0,
+                    }
+                ],
+                "frame_videos": [{"video": "video"}],
+                "reference_images": ["string"],
+                "reference_videos": [{"video": "video"}],
+                "source_video": {"video": "video"},
+            },
             negative_prompt="negative_prompt",
             output_format="MP4",
             output_quality=0,
             prompt="x",
+            ratio="ratio",
             reference_images=["string"],
+            resolution="resolution",
             seconds="seconds",
             seed=0,
             steps=10,