Skip to content

Releases: agno-agi/agno

v2.0.10

25 Sep 21:35
@ysolanky ysolanky
v2.0.10
2eefefa
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.10 Latest
Latest

Changelog

Improvements:

  • Session State: Adds a new flag overwrite_db_session_state to overwrite the session state persisted in the database

Bug Fixes:

  • AWS Claude Model Class: Fixes a bug with Tool call parsing

What's Changed

Full Changelog: v2.0.9...v2.0.10

Contributors

manuhortet, kausmeows, and ysolanky
Assets 2
Loading

v1.8.4

25 Sep 18:39
@ysolanky ysolanky
1.8.4
107ab20
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v1.8.4

Changelog

Bug Fixes:

  • Aws Claude Model Class: Fixed a regression caused in 1.8.3 leading to tool call failure for AWS Claude Model Class

What's Changed

Full Changelog: v1.8.3...v1.8.4

Contributors

kausmeows
Loading

v2.0.9

24 Sep 19:48
@ysolanky ysolanky
v2.0.9
9fd3cc4
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.9

Changelog

New Features:

  • MCP Toolbox: new Toolkit to provide your Agents with access to Google’s MCP Toolbox for Databases.
  • Bulk DB writing: all DB implementations now support writing multiple Sessions and Memories with one DB call.
  • Ollama Cloud: Added support for Ollama cloud.
  • Handle session_state: the AgentOS run endpoints now support receiving the session_state and dependencies parameters

Improvements:

  • Ollama Cloud: Ollama Model class now supports use of Ollama Cloud via an API key
  • AgentOS WhatsApp interface: allow adjusting the WhatsApp version to use with the interface.
  • Workflows support for Slack interface: Workflows can now be used in AgentOS together with the Slack interface.
  • Scrape Graph Toolkit: New methods added to the Scrape Graph Toolkit.
  • Raise on Storage errors: Raise any critical errors that happen when reading or writing from a database, instead of logging them.
  • V2 Migration script: batch operations to support and speed up migrating large tables.

Bug Fixes:

  • AgentOS chat: Fixed a bug where user messages would be displayed in the wrong order.
  • V2 Migration script: Add support for schemas pertaining to older Agno versions.
  • HuggingFace Model Class: Fixed a bug with tool call streaming for HF model class

What's Changed

New Contributors

Full Changelog: v2.0.8...v2.0.9

Contributors

b41sh, manuhortet, and 9 other contributors
Loading

v1.8.3

24 Sep 15:10
@ysolanky ysolanky
v1.8.3
1893526
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v1.8.3

Changelog

Bug Fixes:

  • Anthropic Claude: Removes legacy tool call formatting

What's Changed

Full Changelog: v1.8.2...v1.8.3

Contributors

ysolanky
Loading

v2.0.8

22 Sep 18:09
@ysolanky ysolanky
v2.0.8
83659ee
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.8

Changelog

New Features:

  • CometAPI: Added support for CometAPI as a model provider.

Improvements:

  • Access dependencies on tool: dependencies are now available on custom tools as a built-in argument. See an example.
  • V1 → V2 Migration Script: Updated the migration script by introducing batching. This makes it significantly more performant for large databases.
  • Support multiple text_contents on add_contents() in Knowledge.
  • MultiMCPTools: Added allow_partial_failure flag to MultiMCPTools toolkit

Bug Fixes:

  • Dependencies in Teams: Ensure dependencies are correctly passed through to member agents to reference.
  • Tool call limit on default reasoning agent: Ensures that a tool_call_limit can be passed to the default reasoning agent, avoiding potential costly tool call loops.
  • Fix string encoding: Addresses issue where string encoding broke when adding Chinese string content. Encoding now reliably works via utf-8 encoding.
  • Fix run_input not being populated in session runs endpoint: The /sessions/<session_id>/runs endpoint does not return run_input in case of error as it references it from step_executor_runs and its empty if the workflow hasn't run yet, so in this case we fallback to directly get it from input.
  • Fix State Precedence for Agents/Teams: Fixed issues related to state from the database that would override state passed on agent.run.
  • Fix early stop in workflow to return correct last output: early stop functionality in workflow was not updating the last output properly for non stream cases
  • Fixed OpenAIResponses to remove duplicate messages for reasoning models:
    • The OpenAIResponses API has the (default enabled) feature that they store the conversation server-side and we send a previous_response_id with any followup questions.
    • Agno's OpenAIResponses class implements this feature by looking through the message history, finding the most recent assistant response, and using that message ID as previous_response_id. This part works. However, we were then not deleting the prior messages from the request it sends to OpenAI. As a result OpenAI concatenates its own message history with the message history sent by Agno, resulting in an exponential number of repeated messages as the conversation continues.

What's Changed

New Contributors

Full Changelog: v2.0.7...v2.0.8

Contributors

KeraStudios, jtalmi, and 13 other contributors
Loading

v2.0.7

18 Sep 15:55
@dirkbrnd dirkbrnd
v2.0.7
7693d11
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.7

Changelog

New Features:

  • LlamaCpp Model: Added a new LlamaCpp Model class for Llama CPP support!

Improvements:

  • AgentOS CORS Middleware: AgentOS now automatically merges existing CORSMiddleware allow_origin lists when a custom FastAPI app is provided, with what is required for the AgentOS UI to function correctly.

Bug Fixes:

  • input_schema in UI breaking workflows: For some workflow configs where the steps are not wrapped in Step class, the UI does not show the input_schema form for them. This has been resolved.
  • Improve functionality of send_media_to_model flag:
    • we currently use the send_media_to_model flag for the case where the input image to agent is to be accessed by a tool and should not be available to the model.
    • In this use-case if we use a tool to generate an image (e.g. DallE tools) and using a non-multimodal model, we then append the tool generated media now as a user message to model. When send_media_to_model=False we now exclude these results when sending messages to the model.

What's Changed

New Contributors

Full Changelog: v2.0.6...v2.0.7

Contributors

guodage, adiberk, and 6 other contributors
Loading

v2.0.6

18 Sep 01:07
@ysolanky ysolanky
v2.0.6
29a40a5
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.6

Changelog

New Features:

  • Nexus Model Class: Adds a Model class for Nexus Router
  • File Generation Tools: Added FileGenerationTools to give agents the ability to generate file artifacts! PDFs, CSVs, JSON and TXT. See the docs for more information.

Improvements:

  • Store Workflow Input in DB: The main workflow input is also now stored in WorkflowRunOutput and correspondingly in DB.

Bug Fixes:

  • E2B Tools Parsing: Fixed issues with parsing bytes results.
  • ChromaDB Metadata: Fixed issues related to complex metadata schemas and ChromaDb.
  • Gemini Schema Parsing: Fixed parsing of MCP tool schemas with nullable types and empty item definitions along with union type fields for structured output.
  • File upload in AgentOS chat page: Fix breaking file upload like pdfs in chat page
  • Enabling search_session_history should be user specific: fix on ****enabling search_session_history=True we are not fetching user specific sessions.
  • LanceDB: Fixed a bug with LanceDB metadata update function

What's Changed

New Contributors

Full Changelog: v2.0.5...v2.0.6

Contributors

jonemo, guyrosin, and 10 other contributors
Loading

v2.0.5

17 Sep 02:44
@ysolanky ysolanky
v2.0.5
795d413
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.5

Changelog

Improvements:

  • AgentOS Home: Added route for / to avoid 404 errors.
  • Ag-UI: Fixed Ag-UI <> AgentOS integration.
  • AgentOS with Custom FastAPI: Extended further support for custom FastAPI apps with AgentOS, where existing routes that clash with AgentOS routes are overwritten by default, but can be switched off with replace_routes=False .
  • Migration Script Updates: Updated the v1 → v2 migration script to support parsing for metrics and migrations for MongoDB. See the full migration guide here
  • OpenAI Responses: Added accurate metric tracking for OpenAI Responses

What's Changed

New Contributors

Full Changelog: v2.0.4...v2.0.5

Contributors

anch0vy, kyleaton, and 13 other contributors
Loading

v2.0.4

12 Sep 20:02
@dirkbrnd dirkbrnd
v2.0.4
b3db3bb
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.4

Changelog

New Features:

  • TypedDict support in input_schema for agent/team/workflow with structured input: Along with Pydantic support, TypedDict support is also available.

  • Added SiliconFlow : New model class added for Silicon Flow.

  • session_state as param in custom function step for workflow:
    You can now do-

    # on the function of the custom python function step for a workflow
def custom_function_step(step_input: StepInput, session_state):
    session_state["test"] = test_1 # updates the workflow session state

Improvements:

  • MCP support on AgentOS evals: all evals now support async tools, as MCP.
  • extra_body on OpenAI: Added extra_body parameter on OpenAIChat and OpenAILike.
  • Knowledge Content Resolution: Rehydration of contentsdb rows for existing vectors

Bug Fixes:

  • Dependencies: Resolved issues on Agent/Team for dependency resolution.
  • Agno Infra: Fixed issues related to ag infra up and ag infra down when run outside of the project.
  • Workflows session serialization: handle events serialization like in agent/team.
  • Website reader url handling improved

What's Changed

New Contributors

Full Changelog: v2.0.3...v2.0.4

Contributors

falmanna, manuhortet, and 11 other contributors
Loading

v2.0.3

10 Sep 21:02
@ysolanky ysolanky
v2.0.3
c6d8224
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.3

Changelog

New Features:

  • WorkflowTools: Added WorkflowTools as a tool for agents and teams that allows the agent/team to run a workflow. Also has reasoning support for think/analyze.
  • MemoryTools: Added MemoryTools as a tool for agents and teams to add, update, delete memories. Also has reasoning support for think/analyze.
  • Gemini TTS: Added support for Gemini Text to Speech.
  • Added support for encoding to be passed on reader class.

Improvements:

  • AgentOS kwargs on runs: Added support for additional arguments on running an agent/team/workflow via AgentOS API.

Bug Fixes:

  • Cerebras: Fixed issues with parallel_tool_calls in Cerebras model.
  • OpenAIResponses: Fixed a bug with store=False in OpenAIResponses class.
  • AgentOS Slack Interface: the events endpoint was wrongly exposed on slack/slack/events, it is now exposed on slack/events
  • AgentOS WhatsApp Interface: fixed a bug where some messages wouldn’t be sent.
  • Team MemoryManager: Added memory_manager parameter to Teams.
  • FixedSizeChunking: Default chunk_size increased to 5000.
  • Parser Model output in print_response_stream: In streaming case the structured output from parser model was not being displayed json formatted in cli.

What's Changed

New Contributors

Full Changelog: v2.0.2...v2.0.3

Contributors

whitfin, manuhortet, and 10 other contributors
Loading