Token limited model context is currently broken because it is importing
from extensions.
This fix removed the imports and updated the model context
implementation to use model client directly.
In the future, the model client's token counting should cache results
from model API to provide accurate counting.
Resolves#5745
Also made sure to log LLMCallEvent from all builtin model clients, and
added unit test for coverage.
---------
Co-authored-by: Ryan Sweet <rysweet@microsoft.com>
Co-authored-by: Victor Dibia <victordibia@microsoft.com>
Fixes#4821 by adding a `close()` method to all clients.
Additionally:
* The m1 CLI is updated to close the client before exiting.
* The playwrightcontroller is updated to suppress some other unrelated
chatty warnings (e.g,, produced by markitdown when encountering
conversions that require external utilities)
AutoGen was passing raw dictionaries to functions instead of
constructing Pydantic model or dataclass instances. If a tool function’s
parameter was a Pydantic BaseModel or a dataclass, the function would
receive a dict and likely throw an error or behave incorrectly (since it
expected an object of that type).
This PR addresses problem in AutoGen where tool functions expecting
structured inputs (Pydantic models or dataclasses) were receiving raw
dictionaries. It ensures that structured inputs are automatically
validated and instantiated before function calls. Complete details are
in Issue #5736
[Reproducible Example Code - Failing
Case](https://colab.research.google.com/drive/1hgoP-cGdSZ1-OqQLpwYmlmcExgftDqlO?usp=sharing)
<!-- Please give a short summary of the change and the problem this
solves. -->
## Changes Made:
- Inspect function signatures for Pydantic BaseModel and dataclass
annotations.
- Convert input dictionaries into properly instantiated objects using
BaseModel.model_validate() for Pydantic models or standard instantiation
for dataclasses.
- Raise descriptive errors when validation or instantiation fails.
- Unit tests have been added to cover all scenarios
Now structured inputs are automatically validated and instantiated
before function calls.
- **Updated Conversion Logic:**
In the `run()` method, we now inspect the function’s signature and
convert input dictionaries to structured objects. For parameters
annotated with a Pydantic model, we use `model_validate()` to create an
instance; for those annotated with a dataclass, we instantiate the
object using the dataclass constructor. For example:
```python
# Get the function signature.
sig = inspect.signature(self._func)
raw_kwargs = args.model_dump()
kwargs = {}
# Iterate over the parameters expected by the function.
for name, param in sig.parameters.items():
if name in raw_kwargs:
expected_type = param.annotation
value = raw_kwargs[name]
# If expected type is a subclass of BaseModel, perform conversion.
if inspect.isclass(expected_type) and issubclass(expected_type,
BaseModel):
try:
kwargs[name] = expected_type.model_validate(value)
except ValidationError as e:
raise ValueError(
f"Error validating parameter '{name}' for function
'{self._func.__name__}': {e}"
) from e
# If it's a dataclass, instantiate it.
elif is_dataclass(expected_type):
try:
cls = expected_type if isinstance(expected_type, type) else
type(expected_type)
kwargs[name] = cls(**value)
except Exception as e:
raise ValueError(
f"Error instantiating dataclass parameter '{name}' for function
'{self._func.__name__}': {e}"
) from e
else:
kwargs[name] = value
```
- **Error Handling Improvements:**
Conversion steps are wrapped in try/except blocks to raise descriptive
errors when instantiation fails, aiding in debugging invalid inputs.
- **Testing:**
Unit tests have been added to simulate tool calls (e.g., an `add` tool)
to ensure that with input like:
```json
{"input": {"x": 2, "y": 3}}
```
The tool function receives an instance of the expected type and returns
the correct result.
## Related issue number
Closes#5736
## Checks
- [x] I've included any doc changes needed for
<https://microsoft.github.io/autogen/>. See
<https://github.com/microsoft/autogen/blob/main/CONTRIBUTING.md> to
build and test documentation locally.
- [x] I've added tests (if relevant) corresponding to the changes
introduced in this PR.
- [x] I've made sure all auto checks have passed.
<!-- Thank you for your contribution! Please review
https://microsoft.github.io/autogen/docs/Contribute before opening a
pull request. -->
<!-- Please add a reviewer to the assignee section when you create a PR.
If you don't have the access to it, we will shortly find a reviewer and
assign them to your PR. -->
## Why are these changes needed?
<!-- Please give a short summary of the change and the problem this
solves. -->
The PR introduces two changes.
The first change is adding a name attribute to
`FunctionExecutionResult`. The motivation is that semantic kernel
requires it for their function result interface and it seemed like a
easy modification as `FunctionExecutionResult` is always created in the
context of a `FunctionCall` which will contain the name. I'm unsure if
there was a motivation to keep it out but this change makes it easier to
trace which tool the result refers to and also increases api
compatibility with SK.
The second change is an update to how messages are mapped from autogen
to semantic kernel, which includes an update/fix in the processing of
function results.
## Related issue number
<!-- For example: "Closes #1234" -->
Related to #5675 but wont fix the underlying issue of anthropic
requiring tools during AssistantAgent reflection.
## Checks
- [ ] I've included any doc changes needed for
<https://microsoft.github.io/autogen/>. See
<https://github.com/microsoft/autogen/blob/main/CONTRIBUTING.md> to
build and test documentation locally.
- [ ] I've added tests (if relevant) corresponding to the changes
introduced in this PR.
- [ ] I've made sure all auto checks have passed.
---------
Co-authored-by: Leonardo Pinheiro <lpinheiro@microsoft.com>
Closes#4904
Does not change default behavior in core.
In agentchat, this change will mean that exceptions that used to be
ignored and result in bugs like the group chat stopping are now reported
out to the user application.
---------
Co-authored-by: Ben Constable <benconstable@microsoft.com>
Co-authored-by: Ryan Sweet <rysweet@microsoft.com>
<!-- Thank you for your contribution! Please review
https://microsoft.github.io/autogen/docs/Contribute before opening a
pull request. -->
<!-- Please add a reviewer to the assignee section when you create a PR.
If you don't have the access to it, we will shortly find a reviewer and
assign them to your PR. -->
## Why are these changes needed?
Currently the way to accomplish RAG behavior with agent chat,
specifically assistant agents is with the memory interface, however
there is no way to configure it via the declarative API.
<!-- Please give a short summary of the change and the problem this
solves. -->
## Related issue number
<!-- For example: "Closes #1234" -->
## Checks
- [ ] I've included any doc changes needed for
https://microsoft.github.io/autogen/. See
https://microsoft.github.io/autogen/docs/Contribute#documentation to
build and test documentation locally.
- [ ] I've added tests (if relevant) corresponding to the changes
introduced in this PR.
- [ ] I've made sure all auto checks have passed.
---------
Co-authored-by: Victor Dibia <chuvidi2003@gmail.com>
<!-- Thank you for your contribution! Please review
https://microsoft.github.io/autogen/docs/Contribute before opening a
pull request. -->
<!-- Please add a reviewer to the assignee section when you create a PR.
If you don't have the access to it, we will shortly find a reviewer and
assign them to your PR. -->
## Why are these changes needed?
FunctionTool supports passing in a partial
## Related issue number
Closes#5151
## Checks
- [x] I've included any doc changes needed for
https://microsoft.github.io/autogen/. See
https://microsoft.github.io/autogen/docs/Contribute#documentation to
build and test documentation locally.
- [x] I've added tests (if relevant) corresponding to the changes
introduced in this PR.
- [x] I've made sure all auto checks have passed.
<!-- Thank you for your contribution! Please review
https://microsoft.github.io/autogen/docs/Contribute before opening a
pull request. -->
<!-- Please add a reviewer to the assignee section when you create a PR.
If you don't have the access to it, we will shortly find a reviewer and
assign them to your PR. -->
## Why are these changes needed?
It is currently hard to add a description to a component (defaults to
None also) .. you have to call super.dump() modify and return. This PR
makes the experience better.
- allows you specify `component_description` and `component_label` as an
optional class var. label is an optional human readable name for the the
component.
- will use component_description if provided int he description field
when dumped if there is no description, will use the first line of class
docstring. Takes advantage of all the good practices we have in writing
good docstrings. label defaults to component type.
For example
```python
model_client=OpenAIChatCompletionClient( model="gpt-4o-2024-08-06" )
config = model_client.dump_component()
print(config.model_dump_json())
```
Note the description field below is no longer None and there is a label
```python
{
"provider": "autogen_ext.models.openai.OpenAIChatCompletionClient",
"component_type": "model",
"version": 1,
"component_version": 1,
"description": "Chat completion client for OpenAI hosted models.",
"label": "OpenAIChatCompletionClient",
"config": { "model": "gpt-4o-2024-08-06" }
}
```
<!-- Please give a short summary of the change and the problem this
solves. -->
## Related issue number
<!-- For example: "Closes #1234" -->
None, felt faster to fix.
## Checks
- [x] I've included any doc changes needed for
https://microsoft.github.io/autogen/. See
https://microsoft.github.io/autogen/docs/Contribute#documentation to
build and test documentation locally.
- [x] I've added tests (if relevant) corresponding to the changes
introduced in this PR.
- [x] I've made sure all auto checks have passed.
* initial pass on making group chats declarative
* update group chat tests
* update impl to include participant serialization for all teams
* v1 making soc declarative
* update memory test
* update chatagent and team base classes
* update serialization doc notebook
* fomating updates
* vi1 for declarative tools
* make functtools declarative
* add tests
* update imports
* update formatting
* move tests, format fixes
* format updates
* update test
* add user warning to _from_config
* add warning on load_component to docs
---------
Co-authored-by: Ryan Sweet <rysweet@microsoft.com>
* Add ChatCompletionCache along with AbstractStore for caching completions
* Addressing comments
* Improve interface for cachestore
* Improve documentation & revert protocol
* Make cache store typed, and improve docs
* remove unnecessary casts
* initial base memroy impl
* update, add example with chromadb
* include mimetype consideration
* add transform method
* update to address feedback, will update after 4681 is merged
* update memory impl,
* remove chroma db, typing fixes
* format, add test
* update uv lock
* update docs
* format updates
* update notebook
* add memoryqueryevent message, yield message for observability.
* minor fixes, make score optional/none
* Update python/packages/autogen-agentchat/src/autogen_agentchat/agents/_assistant_agent.py
Co-authored-by: Eric Zhu <ekzhu@users.noreply.github.com>
* update tests to improve cov
* refactor, move memory to core.
* format fixxes
* format updates
* format updates
* fix azure notebook import, other fixes
* update notebook, support str query in Memory protocol
* update test
* update cells
* add specific extensible return types to memory query and update_context
---------
Co-authored-by: Eric Zhu <ekzhu@users.noreply.github.com>
* Split apart component infra to allow for abstract class integration
* fix is_component_class check
* make is_ functions type guards
* Simplify component creation
* undo changes
* Format
* Instantiate and call activate/deactivate on agents
* autoformatting
* remove activate. Rename deactivate to close
* remove unneeded import
* create close fn in runtime
* change runtime close behavior
* uv.lock
---------
Co-authored-by: Jack Gerrits <jackgerrits@users.noreply.github.com>
* Decouple model_context from AssistantAgent
* add UnboundedBufferedChatCompletionContext to mimic pervious model_context behaviour on AssistantAgent
* moving unbounded buffered chat to a different file
* fix model_context assertions in test_group_chat
* Refactor model context, introduce states
* fixes
* update
---------
Co-authored-by: aditya.kurniawan <aditya.kurniawan@core42.ai>
Co-authored-by: Eric Zhu <ekzhu@users.noreply.github.com>
Co-authored-by: Victor Dibia <victordibia@microsoft.com>
* Move LLMUsageTracker to a cookbook
* rename, add to index
* fmt and remove test
* Update python/packages/autogen-core/docs/src/user-guide/core-user-guide/cookbook/llm-usage-logger.ipynb
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Fix missing quotation marks in notebook
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Eric Zhu