agno 2.0.0rc1__py3-none-any.whl → 2.0.1__py3-none-any.whl

agno/tools/brightdata.py

@@ -5,7 +5,7 @@ from typing import Any, Dict, List, Optional
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact
+from agno.media import Image
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
 from agno.utils.log import log_debug, log_error, log_info
@@ -158,8 +158,8 @@ class BrightDataTools(Toolkit):
 
             media_id = str(uuid4())
 
-            # Create ImageArtifact for the screenshot
-            image_artifact = ImageArtifact(
+            # Create Image for the screenshot
+            image_artifact = Image(
                 id=media_id,
                 content=base64_encoded_image.encode("utf-8"),
                 mime_type="image/png",

agno/tools/cartesia.py

@@ -1,11 +1,10 @@
 import json
-from base64 import b64encode
 from os import getenv
 from typing import Any, Dict, List, Optional, Union
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import AudioArtifact
+from agno.media import Audio
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -170,12 +169,11 @@ class CartesiaTools(Toolkit):
 
             audio_iterator = self.client.tts.bytes(**params)
             audio_data = b"".join(chunk for chunk in audio_iterator)
-            base64_audio = b64encode(audio_data).decode("utf-8")
 
             # Create AudioArtifact
-            audio_artifact = AudioArtifact(
+            audio_artifact = Audio(
                 id=str(uuid4()),
-                base64_audio=base64_audio,
+                content=audio_data,
                 mime_type=mime_type,  # Hardcoded to audio/mpeg
             )
 

agno/tools/dalle.py

@@ -3,7 +3,7 @@ from typing import Any, List, Literal, Optional, Union
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact
+from agno.media import Image
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -94,10 +94,13 @@ class DalleTools(Toolkit):
             if response.data:
                 for img in response.data:
                     if img.url:
-                        image_artifact = ImageArtifact(
-                            id=str(uuid4()), url=img.url, original_prompt=prompt, revised_prompt=img.revised_prompt
+                        image = Image(
+                            id=str(uuid4()),
+                            url=img.url,
+                            original_prompt=prompt,
+                            revised_prompt=img.revised_prompt,
                         )
-                        generated_images.append(image_artifact)
+                        generated_images.append(image)
                         response_str += f"Image has been generated at the URL {img.url}\n"
 
             return ToolResult(

agno/tools/desi_vocal.py

@@ -5,7 +5,7 @@ from uuid import uuid4
 import requests
 
 from agno.agent import Agent
-from agno.media import AudioArtifact
+from agno.media import Audio
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -97,7 +97,7 @@ class DesiVocalTools(Toolkit):
             response_json = response.json()
             audio_url = response_json["s3_path"]
 
-            audio_artifact = AudioArtifact(id=str(uuid4()), url=audio_url)
+            audio_artifact = Audio(id=str(uuid4()), url=audio_url)
 
             return ToolResult(
                 content=f"Audio generated successfully: {audio_url}",

agno/tools/e2b.py

@@ -8,7 +8,7 @@ from typing import Any, Callable, Dict, List, Optional, Union
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact
+from agno.media import Image
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -161,7 +161,7 @@ class E2BTools(Toolkit):
         self, agent: Union[Agent, Team], result_index: int = 0, output_path: Optional[str] = None
     ) -> ToolResult:
         """
-        Add a PNG image result from the last code execution as an ImageArtifact.
+        Add a PNG image result from the last code execution as an Image object.
 
         Args:
             agent: The agent to add the image artifact to
@@ -205,9 +205,9 @@ class E2BTools(Toolkit):
             # Generate a file:// URL for the temp file
             file_url = f"file://{temp_path}"
 
-            # Create ImageArtifact
+            # Create Image object
             image_id = str(uuid4())
-            image_artifact = ImageArtifact(
+            image_artifact = Image(
                 id=image_id, url=file_url, original_prompt=f"Generated from code execution result {result_index}"
             )
 
@@ -292,9 +292,9 @@ class E2BTools(Toolkit):
                 # Generate a file:// URL for the temp file
                 file_url = f"file://{temp_path}"
 
-                # Create ImageArtifact
+                # Create Image object
                 image_id = str(uuid4())
-                image_artifact = ImageArtifact(
+                image_artifact = Image(
                     id=image_id, url=file_url, original_prompt=f"Interactive {chart_type} chart from code execution"
                 )
 

agno/tools/eleven_labs.py

@@ -6,7 +6,7 @@ from typing import Any, Iterator, List, Literal, Optional, Union
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import AudioArtifact
+from agno.media import Audio
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -147,7 +147,7 @@ class ElevenLabsTools(Toolkit):
             base64_audio = self._process_audio(audio_generator)
 
             # Create AudioArtifact
-            audio_artifact = AudioArtifact(
+            audio_artifact = Audio(
                 id=str(uuid4()),
                 base64_audio=base64_audio,
                 mime_type="audio/mpeg",
@@ -182,7 +182,7 @@ class ElevenLabsTools(Toolkit):
             base64_audio = self._process_audio(audio_generator)
 
             # Create AudioArtifact
-            audio_artifact = AudioArtifact(
+            audio_artifact = Audio(
                 id=str(uuid4()),
                 base64_audio=base64_audio,
                 mime_type="audio/mpeg",

agno/tools/fal.py

@@ -7,7 +7,7 @@ from typing import Optional, Union
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact, VideoArtifact
+from agno.media import Image, Video
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -72,14 +72,14 @@ class FalTools(Toolkit):
 
             if "image" in result:
                 url = result.get("image", {}).get("url", "")
-                image_artifact = ImageArtifact(
+                image_artifact = Image(
                     id=media_id,
                     url=url,
                 )
                 return ToolResult(content=f"Image generated successfully at {url}", images=[image_artifact])
             elif "video" in result:
                 url = result.get("video", {}).get("url", "")
-                video_artifact = VideoArtifact(
+                video_artifact = Video(
                     id=media_id,
                     url=url,
                 )
@@ -115,7 +115,7 @@ class FalTools(Toolkit):
             )
             url = result.get("images", [{}])[0].get("url", "")
             media_id = str(uuid4())
-            image_artifact = ImageArtifact(
+            image_artifact = Image(
                 id=media_id,
                 url=url,
             )

agno/tools/function.py

@@ -8,7 +8,7 @@ from packaging.version import Version
 from pydantic import BaseModel, Field, validate_call
 
 from agno.exceptions import AgentRunException
-from agno.media import Audio, AudioArtifact, File, Image, ImageArtifact, Video, VideoArtifact
+from agno.media import Audio, File, Image, Video
 from agno.utils.log import log_debug, log_error, log_exception, log_warning
 
 T = TypeVar("T")
@@ -482,9 +482,9 @@ class FunctionExecutionResult(BaseModel):
     updated_session_state: Optional[Dict[str, Any]] = None
 
     # New fields for media artifacts
-    images: Optional[List[ImageArtifact]] = None
-    videos: Optional[List[VideoArtifact]] = None
-    audios: Optional[List[AudioArtifact]] = None
+    images: Optional[List[Image]] = None
+    videos: Optional[List[Video]] = None
+    audios: Optional[List[Audio]] = None
 
 
 class FunctionCall(BaseModel):
@@ -962,6 +962,6 @@ class ToolResult(BaseModel):
     """Result from a tool that can include media artifacts."""
 
     content: str
-    images: Optional[List[ImageArtifact]] = None
-    videos: Optional[List[VideoArtifact]] = None
-    audios: Optional[List[AudioArtifact]] = None
+    images: Optional[List[Image]] = None
+    videos: Optional[List[Video]] = None
+    audios: Optional[List[Audio]] = None

agno/tools/giphy.py

@@ -5,7 +5,7 @@ from typing import Any, List, Optional, Union
 import httpx
 
 from agno.agent import Agent
-from agno.media import ImageArtifact
+from agno.media import Image
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -77,7 +77,7 @@ class GiphyTools(Toolkit):
                 gif_urls.append(gif_url)
 
                 # Create ImageArtifact for the GIF
-                image_artifact = ImageArtifact(id=media_id, url=gif_url, alt_text=alt_text, revised_prompt=query)
+                image_artifact = Image(id=media_id, url=gif_url, alt_text=alt_text, revised_prompt=query)
                 image_artifacts.append(image_artifact)
 
             if image_artifacts:

agno/tools/lumalab.py

@@ -4,7 +4,7 @@ from os import getenv
 from typing import Any, Dict, List, Literal, Optional, TypedDict
 
 from agno.agent import Agent
-from agno.media import VideoArtifact
+from agno.media import Video
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
 from agno.utils.log import log_info, logger
@@ -109,7 +109,7 @@ class LumaLabTools(Toolkit):
                 if generation.state == "completed" and generation.assets:
                     video_url = generation.assets.video
                     if video_url:
-                        video_artifact = VideoArtifact(id=video_id, url=video_url, eta="completed")
+                        video_artifact = Video(id=video_id, url=video_url, eta="completed")
                         return ToolResult(
                             content=f"Video generated successfully: {video_url}",
                             videos=[video_artifact],
@@ -164,7 +164,7 @@ class LumaLabTools(Toolkit):
                 if generation.state == "completed" and generation.assets:
                     video_url = generation.assets.video
                     if video_url:
-                        video_artifact = VideoArtifact(id=video_id, url=video_url, state="completed")
+                        video_artifact = Video(id=video_id, url=video_url, state="completed")
                         return ToolResult(
                             content=f"Video generated successfully: {video_url}",
                             videos=[video_artifact],

agno/tools/mcp.py

@@ -365,7 +365,7 @@ class MultiMCPTools(Toolkit):
         *,
         env: Optional[dict[str, str]] = None,
         server_params_list: Optional[
-            List[Union[SSEClientParams, StdioServerParameters, StreamableHTTPClientParams]]
+            list[Union[SSEClientParams, StdioServerParameters, StreamableHTTPClientParams]]
         ] = None,
         timeout_seconds: int = 5,
         client=None,
@@ -531,7 +531,6 @@ class MultiMCPTools(Toolkit):
                 session = await self._async_exit_stack.enter_async_context(ClientSession(read, write))
                 self._active_contexts.append(session)
                 await self.initialize(session)
-
             # Handle Streamable HTTP connections
             elif isinstance(server_params, StreamableHTTPClientParams):
                 client_connection = await self._async_exit_stack.enter_async_context(

agno/tools/models/azure_openai.py

@@ -5,7 +5,7 @@ from uuid import uuid4
 from requests import post
 
 from agno.agent import Agent
-from agno.media import ImageArtifact
+from agno.media import Image
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
 from agno.utils.log import log_debug, logger
@@ -170,7 +170,7 @@ class AzureOpenAITools(Toolkit):
                 revised_prompt = img.get("revised_prompt")
 
                 # Create ImageArtifact with URL
-                image_artifact = ImageArtifact(
+                image_artifact = Image(
                     id=str(uuid4()), url=image_url, original_prompt=prompt, revised_prompt=revised_prompt
                 )
                 generated_images.append(image_artifact)

agno/tools/models/gemini.py

@@ -5,7 +5,7 @@ from typing import Any, Optional
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact, VideoArtifact
+from agno.media import Image, Video
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
 from agno.utils.log import log_debug, log_error, log_info
@@ -109,7 +109,7 @@ class GeminiTools(Toolkit):
                 media_id = str(uuid4())
 
                 # Create ImageArtifact with raw bytes (not base64 encoded)
-                image_artifact = ImageArtifact(
+                image_artifact = Image(
                     id=media_id,
                     content=image_bytes,
                     original_prompt=prompt,
@@ -181,7 +181,7 @@ class GeminiTools(Toolkit):
                 media_id = str(uuid4())
 
                 # Create VideoArtifact with base64 encoded content
-                video_artifact = VideoArtifact(
+                video_artifact = Video(
                     id=media_id,
                     content=base64.b64encode(generated_video.video_bytes).decode("utf-8"),
                     original_prompt=prompt,

agno/tools/models/groq.py

@@ -1,11 +1,10 @@
-import base64
 from os import getenv
 from pathlib import Path
 from typing import Any, List, Optional
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import AudioArtifact
+from agno.media import Audio
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
 from agno.utils.log import log_debug, log_error
@@ -143,12 +142,11 @@ class GroqTools(Toolkit):
             log_debug(f"Groq TTS Response: {response}")
 
             audio_data: bytes = response.read()
-            base64_encoded_audio = base64.b64encode(audio_data).decode("utf-8")
 
             media_id = str(uuid4())
-            audio_artifact = AudioArtifact(
+            audio_artifact = Audio(
                 id=media_id,
-                base64_audio=base64_encoded_audio,
+                content=audio_data,
                 mime_type="audio/wav",
             )
 

agno/tools/models/nebius.py

@@ -4,7 +4,7 @@ from typing import Optional
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact
+from agno.media import Image
 from agno.models.nebius import Nebius
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -108,7 +108,7 @@ class NebiusTools(Toolkit):
                 media_id = str(uuid4())
 
                 # Create ImageArtifact with raw bytes
-                image_artifact = ImageArtifact(
+                image_artifact = Image(
                     id=media_id, content=image_content_bytes, mime_type="image/png", original_prompt=prompt
                 )
 

agno/tools/models_labs.py

@@ -5,7 +5,7 @@ from typing import Any, Dict, List, Optional, Union
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import AudioArtifact, ImageArtifact, VideoArtifact
+from agno.media import Audio, Image, Video
 from agno.models.response import FileType
 from agno.team import Team
 from agno.tools import Toolkit
@@ -89,20 +89,20 @@ class ModelsLabTools(Toolkit):
 
     def _create_media_artifacts(self, media_id: str, media_url: str, eta: Optional[str] = None) -> Dict[str, List]:
         """Create appropriate media artifacts based on file type."""
-        artifacts: Dict[str, List[Union[ImageArtifact, VideoArtifact, AudioArtifact]]] = {
+        artifacts: Dict[str, List[Union[Image, Video, Audio]]] = {
             "images": [],
             "videos": [],
             "audios": [],
         }
 
         if self.file_type == FileType.MP4:
-            video_artifact = VideoArtifact(id=str(media_id), url=media_url, eta=str(eta))
+            video_artifact = Video(id=str(media_id), url=media_url, eta=str(eta))
             artifacts["videos"].append(video_artifact)
         elif self.file_type == FileType.GIF:
-            image_artifact = ImageArtifact(id=str(media_id), url=media_url)
+            image_artifact = Image(id=str(media_id), url=media_url)
             artifacts["images"].append(image_artifact)
         elif self.file_type == FileType.MP3:
-            audio_artifact = AudioArtifact(id=str(media_id), url=media_url)
+            audio_artifact = Audio(id=str(media_id), url=media_url)
             artifacts["audios"].append(audio_artifact)
 
         return artifacts

agno/tools/openai.py

@@ -3,7 +3,7 @@ from typing import Any, List, Literal, Optional, Union
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import AudioArtifact, ImageArtifact
+from agno.media import Audio, Image
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -149,7 +149,7 @@ class OpenAITools(Toolkit):
                 image_bytes = base64.b64decode(image_base64)
 
                 # Create ImageArtifact and return in ToolResult
-                image_artifact = ImageArtifact(
+                image_artifact = Image(
                     id=media_id,
                     content=image_bytes,  # ← Store as bytes, not encoded string
                     mime_type="image/png",
@@ -176,8 +176,6 @@ class OpenAITools(Toolkit):
             text_input (str): The text to synthesize into speech.
         """
         try:
-            import base64
-
             response = OpenAIClient(api_key=self.api_key).audio.speech.create(
                 model=self.tts_model,
                 voice=self.tts_voice,
@@ -188,14 +186,11 @@ class OpenAITools(Toolkit):
             # Get raw audio data for artifact creation before potentially saving
             audio_data: bytes = response.content
 
-            # Base64 encode the audio data
-            base64_encoded_audio = base64.b64encode(audio_data).decode("utf-8")
-
             # Create AudioArtifact and return in ToolResult
             media_id = str(uuid4())
-            audio_artifact = AudioArtifact(
+            audio_artifact = Audio(
                 id=media_id,
-                base64_audio=base64_encoded_audio,
+                content=audio_data,
                 mime_type=f"audio/{self.tts_format}",
             )
 

agno/tools/opencv.py

@@ -4,7 +4,7 @@ from typing import Callable, List
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact, VideoArtifact
+from agno.media import Image, Video
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
 from agno.utils.log import log_debug, log_error, log_info
@@ -120,7 +120,7 @@ class OpenCVTools(Toolkit):
                 media_id = str(uuid4())
 
                 # Create ImageArtifact with raw bytes (not base64 encoded)
-                image_artifact = ImageArtifact(
+                image_artifact = Image(
                     id=media_id,
                     content=image_bytes,  # Store as raw bytes
                     original_prompt=prompt,
@@ -292,7 +292,7 @@ class OpenCVTools(Toolkit):
                 media_id = str(uuid4())
 
                 # Create VideoArtifact with base64 encoded content
-                video_artifact = VideoArtifact(
+                video_artifact = Video(
                     id=media_id,
                     content=video_bytes,
                     original_prompt=prompt,

agno/tools/replicate.py

@@ -5,7 +5,7 @@ from urllib.parse import urlparse
 from uuid import uuid4
 
 from agno.agent import Agent
-from agno.media import ImageArtifact, VideoArtifact
+from agno.media import Image, Video
 from agno.team.team import Team
 from agno.tools import Toolkit
 from agno.tools.function import ToolResult
@@ -72,9 +72,9 @@ class ReplicateTools(Toolkit):
                 result_msg, media_artifact = self._parse_output(output)
                 results.append(result_msg)
 
-                if isinstance(media_artifact, ImageArtifact):
+                if isinstance(media_artifact, Image):
                     images.append(media_artifact)
-                elif isinstance(media_artifact, VideoArtifact):
+                elif isinstance(media_artifact, Video):
                     videos.append(media_artifact)
 
             content = "\n".join(results)
@@ -87,7 +87,7 @@ class ReplicateTools(Toolkit):
             logger.error(f"Failed to generate media: {e}")
             return ToolResult(content=f"Error: {e}")
 
-    def _parse_output(self, output: FileOutput) -> Tuple[str, Union[ImageArtifact, VideoArtifact]]:
+    def _parse_output(self, output: FileOutput) -> Tuple[str, Union[Image, Video]]:
         """
         Parse the outputs from the replicate model.
         """
@@ -101,14 +101,14 @@ class ReplicateTools(Toolkit):
         video_extensions = {".mp4", ".mov", ".avi", ".mkv", ".flv", ".wmv", ".webm"}
 
         media_id = str(uuid4())
-        artifact: Union[ImageArtifact, VideoArtifact]
+        artifact: Union[Image, Video]
         media_type: str
 
         if ext in image_extensions:
-            artifact = ImageArtifact(id=media_id, url=output.url)
+            artifact = Image(id=media_id, url=output.url)
             media_type = "image"
         elif ext in video_extensions:
-            artifact = VideoArtifact(id=media_id, url=output.url)
+            artifact = Video(id=media_id, url=output.url)
             media_type = "video"
         else:
             logger.error(f"Unsupported media type with extension '{ext}' for URL: {output.url}")

agno/utils/events.py

@@ -1,6 +1,6 @@
 from typing import Any, List, Optional
 
-from agno.media import AudioResponse, ImageArtifact
+from agno.media import Audio, Image
 from agno.models.message import Citations
 from agno.models.response import ToolExecution
 from agno.reasoning.step import ReasoningStep
@@ -344,8 +344,8 @@ def create_run_output_content_event(
     reasoning_content: Optional[str] = None,
     redacted_reasoning_content: Optional[str] = None,
     citations: Optional[Citations] = None,
-    response_audio: Optional[AudioResponse] = None,
-    image: Optional[ImageArtifact] = None,
+    response_audio: Optional[Audio] = None,
+    image: Optional[Image] = None,
 ) -> RunContentEvent:
     thinking_combined = (reasoning_content or "") + (redacted_reasoning_content or "")
 
@@ -374,8 +374,8 @@ def create_team_run_output_content_event(
     reasoning_content: Optional[str] = None,
     redacted_reasoning_content: Optional[str] = None,
     citations: Optional[Citations] = None,
-    response_audio: Optional[AudioResponse] = None,
-    image: Optional[ImageArtifact] = None,
+    response_audio: Optional[Audio] = None,
+    image: Optional[Image] = None,
 ) -> TeamRunContentEvent:
     thinking_combined = (reasoning_content or "") + (redacted_reasoning_content or "")
 

agno/utils/gemini.py

@@ -19,7 +19,7 @@ def format_image_for_message(image: Image) -> Optional[Dict[str, Any]]:
     # Case 1: Image is a URL
     # Download the image from the URL and add it as base64 encoded data
     if image.url is not None:
-        content_bytes = image.image_url_content
+        content_bytes = image.get_content_bytes()  # type: ignore
         if content_bytes is not None:
             try:
                 import base64