NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
PyPI, #2885 most downloaded on PyPI
The programming language for agentic software.
Latest stable
3.2.0
Last release 10 days ago
12 Nov 2026
Ships on a steady schedule
a new release about every 9 days
Nearly every release is documented
notes for 59 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
2 years old
278 releases · first in 2024
Drag across the ruler to compare, or tap a release to open it.
Support for Agent(team=[]) is still there, but deprecated (see below).
Team in one of 3 modes: “Collaborate”, “Coordinate” or “Route”.debug_mode=True and team.print_response(...) to see it in action.Agent(team=[]) is still there, but deprecated (see below).OpenAILike interface.use_json_mode: bool = False as a parameter of Agent and Team, which in conjunction with response_model=YourModel, is used to indicate whether the agent/team model should be forced to respond in json instead of (now default) structured output. Previous behaviour defaulted to “json-mode”, but since most models now support native structured output, we are now defaulting to native structured output. It is now also much simpler to work with response models, since now only response_model needs to be set. It is not necessary anymore to set structured_output=True to specifically get structured output from the model.WebsiteTools to also update combined knowledgebases.get_message_pairs() fetching incorrect messages.Agent.structured_output has been replaced by Agent.use_json_mode. This will be removed in a future major version release.Agent.team is deprecated with the release of our new Teams implementation. This will be removed in a future major version release.Full Changelog: https://github.com/agno-agi/agno/compare/v.1.1.13...v1.1.14
Your coding agent can read these notes before it upgrades. Set up the MCP server →
OpenAIResponses File Search: Added support for the built-in “File Search” function from OpenAI. This automatically uploads File objects attached to th
File objects attached to the agent prompt.File objects are attached to agent prompts.Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.12...v.1.1.13
Citations: Improved support for capturing, displaying, and storing citations from models, with integration for Gemini and Perplexity.
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.11...v1.1.12
OpenAI Responses: Added a new model implementation that supports OpenAI’s Responses API. This includes support for their “websearch” built-in tool.
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.10...v1.1.11
File Prompts: Introduced a new File type that can be added to prompts and will be sent to the model providers. Only Gemini and Anthropic Claude suppor
File type that can be added to prompts and will be sent to the model providers. Only Gemini and Anthropic Claude supported for now.Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.9...v1.1.10
IBM Watson X: Added support for IBM Watson X as a model provider. Find the docs here.
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.8...v1.1.9
Video File Upload on Playground: You can now upload video files and have a model interpret the video. This only works with certain Gemini models that
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.7...v1.1.8
Audio File Upload on Playground: You can now upload audio files and have a model interpret the audio, do sentiment analysis, provide an audio transcri
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.6...v1.1.7
Claude 3.7 Support: Added support for the latest Claude 3.7 Sonnet model
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.5...v1.1.6
Audio Responses: Agents can now deliver audio responses (both with streaming and non-streaming).
The audio is in the agent.run_response.response_audio.
This only works with OpenAIChat with the gpt-4o-audio-preview model. See their docs for more on how it works. For example
from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.utils.audio import write_audio_to_file
agent = Agent(
model=OpenAIChat(
id="gpt-4o-audio-preview",
modalities=["text", "audio"], # Both text and audio responses are provided.
audio={"voice": "alloy", "format": "wav"},
),
)
agent.print_response(
"Tell me a 5 second story"
)
if agent.run_response.response_audio is not None:
write_audio_to_file(
audio=agent.run_response.response_audio.base64_audio, filename=str(filename)
)
See the audio_conversation_agent cookbook to test it out on the Agent Playground.
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.4...v1.1.5
Gmail Tools: Added get_emails_by_thread and send_email_reply methods to GmailTools.
get_emails_by_thread and send_email_reply methods to GmailTools.openai would be required for chunking even if a model was passed.Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.3...v1.1.4
Gemini Tool-Call History: Fixed an issue where Gemini rejected tool-calls from historic messages.
Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.2...v1.1.3
Reasoning with o3 Models: Reasoning support added for OpenAI’s o3 models.
Reasoning with o3 Models: Reasoning support added for OpenAI’s o3 models.
Gemini embedder update: Updated the GeminiEmbedder to use the new Google’s genai SDK. This update introduces a slight change in the interface:
# Before
embeddings = GeminiEmbedder("models/text-embedding-004").get_embedding(
"The quick brown fox jumps over the lazy dog."
)
# After
embeddings = GeminiEmbedder("text-embedding-004").get_embedding(
"The quick brown fox jumps over the lazy dog."
)
on_bad_vectors as parameters.Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.1...v1.1.2
…audio generation API calls. This is a breaking change, as the parameters for the ModelsLabTools class have changed. The url and fetch_url parameters h…
File / Image Uploads on Agent UI: Agent UI now supports file and image uploads with prompts.
.pdf .csv .txt .docx .json.png .jpeg .jpg .webpFirecrawl Custom API URL: Allowed users to set a custom API URL for Firecrawl.
Updated ModelsLabTools Toolkit Constructor: The constructor in /libs/agno/tools/models_labs.py has been updated to accommodate audio generation API calls. This is a breaking change, as the parameters for the ModelsLabTools class have changed. The url and fetch_url parameters have been removed, and API URLs are now decided based on the file_type provided by the user.
MODELS_LAB_URLS = {
"MP4": "https://modelslab.com/api/v6/video/text2video",
"MP3": "https://modelslab.com/api/v6/voice/music_gen",
"GIF": "https://modelslab.com/api/v6/video/text2video",
}
MODELS_LAB_FETCH_URLS = {
"MP4": "https://modelslab.com/api/v6/video/fetch",
"MP3": "https://modelslab.com/api/v6/voice/fetch",
"GIF": "https://modelslab.com/api/v6/video/fetch",
}
The FileType enum now includes MP3 type:
class FileType(str, Enum):
MP4 = "mp4"
GIF = "gif"
MP3 = "mp3"
MistralEmbedder.Full Changelog: https://github.com/agno-agi/agno/compare/v1.1.0...v1.1.1
Models Refactor: A complete overhaul of our models implementation to improve on performance and to have better feature parity across models.
AwsBedrock.Gemini . This will allow for easier integration of Gemini features in future.exponential_backoff is set to True. <docs>GoogleSearch to GoogleSearchTools for consistency.Gemini implementation directly on the Vertex API has been replaced by the Google SDK implementation of Gemini.Gemini implementation via the OpenAI client has been replaced by the Google SDK implementation of Gemini.OllamaHermes has been removed as the implementation of Ollama was improved.Full Changelog: https://github.com/agno-agi/agno/compare/v1.0.8...v1.1.0
Perplexity Model: We now support [Perplexity](https://www.perplexity.ai/) as a model provider.
name_exists function for LanceDbrun_messages.messages for every run in storageFull Changelog: https://github.com/agno-agi/agno/compare/v1.0.6...v1.0.8
Google Maps Toolkit: Added a rich toolkit for Google Maps that includes business discovery, directions, navigation, geocode locations, nearby places,
Full Changelog: https://github.com/agno-agi/agno/compare/v1.0.5...v1.0.6
Gmail Tools: Add tools for Gmail, including mail search, sending mails, etc.
find_similar to ExaToolsawait agent.aprint_response() and await agent.arun().Full Changelog: https://github.com/agno-agi/agno/compare/v1.0.4...v1.0.5
Fixed a bug with Claude not working with tools that have no params
Full Changelog: https://github.com/agno-agi/agno/compare/v1.0.3...v1.0.4
Add reasoning param by @ashpreetbedi in https://github.com/agno-agi/agno/pull/1964
Full Changelog: https://github.com/agno-agi/agno/compare/v1.0.2...v1.0.3
Model Client Caching: Made all models cache the client instantiation, improving Agno agent instantiation time
TwitterTools to XTools and updated capabilities to be compatible with Twitter API v2.slots=true from the agent dataclass decorator which was not compatible with Python <3.10Full Changelog: https://github.com/agno-agi/agno/compare/v1.0.1...v1.0.2
Enabled caching for Mistral models.
Full Changelog: https://github.com/agno-agi/agno/compare/v1.0.0...v1.0.1
This is a major refactor to be coupled with the launch of Agno.
This is a major refactor to be coupled with the launch of Agno.
phi.model.x → agno.models.x
phi.knowledge_base.x → agno.knowledge.x (applies to all knowledge bases)
phi.document.reader.xxx → agno.document.reader.xxx_reader (applies to all document readers)
All Agno toolkits are now suffixed with Tools. E.g. DuckDuckGo → DuckDuckGoTools
Multi-modal interface updates:
agent.run(images=[]) and agent.print_response(images=[]) is now of type Image
class Image(BaseModel):
url: Optional[str] = None # Remote location for image
filepath: Optional[Union[Path, str]] = None # Absolute local location for image
content: Optional[Any] = None # Actual image bytes content
detail: Optional[str] = None # low, medium, high or auto (per OpenAI spec https://platform.openai.com/docs/guides/vision?lang=node#low-or-high-fidelity-image-understanding)
id: Optional[str] = None
agent.run(audio=[]) and agent.print_response(audio=[]) is now of type Audio
class Audio(BaseModel):
filepath: Optional[Union[Path, str]] = None # Absolute local location for audio
content: Optional[Any] = None # Actual audio bytes content
format: Optional[str] = None
agent.run(video=[]) and agent.print_response(video=[]) is now of type Video
class Video(BaseModel):
filepath: Optional[Union[Path, str]] = None # Absolute local location for video
content: Optional[Any] = None # Actual video bytes content
RunResponse.images is now a list of type ImageArtifact
class ImageArtifact(Media):
id: str
url: str # Remote location for file
alt_text: Optional[str] = None
RunResponse.audio is now a list of type AudioArtifact
class AudioArtifact(Media):
id: str
url: Optional[str] = None # Remote location for file
base64_audio: Optional[str] = None # Base64-encoded audio data
length: Optional[str] = None
mime_type: Optional[str] = None
RunResponse.videos is now a list of type VideoArtifact
class VideoArtifact(Media):
id: str
url: str # Remote location for file
eta: Optional[str] = None
length: Optional[str] = None
RunResponse.response_audio is now of type AudioOutput
class AudioOutput(BaseModel):
id: str
content: str # Base64 encoded
expires_at: int
transcript: str
Models:
Hermes → OllamaHermesAzureOpenAIChat → AzureOpenAICohereChat → CohereDeepSeekChat → DeepSeekGeminiOpenAIChat → GeminiOpenAIHuggingFaceChat → HuggingFaceEmbedders now all take id instead of model as a parameter. For example
db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"
knowledge_base = PDFUrlKnowledgeBase(
urls=["https://phi-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"],
vector_db=PgVector(
table_name="recipes",
db_url=db_url,
embedder=OllamaEmbedder(id="llama3.2", dimensions=3072),
),
)
knowledge_base.load(recreate=True)
Agent Storage class
PgAgentStorage → PostgresAgentStorageSqlAgentStorage → SqliteAgentStorageMongoAgentStorage → MongoDbAgentStorageS2AgentStorage → SingleStoreAgentStorageWorkflow Storage class
SqlWorkflowStorage → SqliteWorkflowStoragePgWorkflowStorage → PostgresWorkflowStorageMongoWorkflowStorage → MongoDbWorkflowStorageKnowledge Base
phi.knowledge.pdf.PDFUrlKnowledgeBase → agno.knowledge.pdf_url.PDFUrlKnowledgeBasephi.knowledge.csv.CSVUrlKnowledgeBase → agno.knowledge.csv_url.CSVUrlKnowledgeBaseReaders
phi.document.reader.arxiv → agno.document.reader.arxiv_readerphi.document.reader.docx → agno.document.reader.docx_readerphi.document.reader.json → agno.document.reader.json_readerphi.document.reader.pdf → agno.document.reader.pdf_readerphi.document.reader.s3.pdf → agno.document.reader.s3.pdf_readerphi.document.reader.s3.text → agno.document.reader.s3.text_readerphi.document.reader.text → agno.document.reader.text_readerphi.document.reader.website → agno.document.reader.website_readerAssistantllmPhiTools toolPythonAgent and DuckDbAgent (this will be brought back in future with more specific agents)similarity_threshold param with threshold paramNothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
One column per month.