KEMBAR78
Add support for AWS Bedrock LLM Provider by yczhang-nv · Pull Request #238 · NVIDIA/NeMo-Agent-Toolkit · GitHub
Skip to content

Conversation

@yczhang-nv
Copy link
Contributor

Description

  • Add support for AWS Bedrock LLM provider. The provider can be accessed by type aws_bedrock, and it is compatible with both langchain and llamaindex
  • Added unit tests for existing combinations of existing LLM providers and LLM frameworks. The unit tests are skipped by default, since each of them needs credential to interact with LLM models, but it should be updated and checked each time we add/modify the LLM provider/client to make sure they are still working
  • Fixed OpenAI + llamaindex LLM client that is not working

Closes AIQ-1213

By Submitting this PR I confirm:

  • I am familiar with the Contributing Guidelines.
  • We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
    • Any contribution which contains commits that are not Signed-Off will not be accepted.
  • When the PR is ready for review, new or existing tests cover these changes.
  • When the PR is ready for review, the documentation is up to date with these changes.

@yczhang-nv yczhang-nv self-assigned this May 9, 2025
@yczhang-nv yczhang-nv added improvement Improvement to existing functionality non-breaking Non-breaking change labels May 9, 2025
@yczhang-nv yczhang-nv force-pushed the yuchen-fix-aiq-1213 branch from 3145f3e to 929f30e Compare May 9, 2025 23:12
dagardner-nv and others added 26 commits May 9, 2025 16:38
* Fix broken anchor tags and external links in examples
* I opened up NVIDIA#176 to prevent this in the future

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - David Gardner (https://github.com/dagardner-nv)

Approvers:
  - Yuchen Zhang (https://github.com/yczhang-nv)

URL: NVIDIA#177
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
* support python 3.11

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* fix doc

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update pyproject.toml

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update uv.lock

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update README to specify Python version requirements

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* resolve comments

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* fix comments

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* remove support for python 3.13

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

---------

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
* Adds temporary transitional packages for users which still have dependencies on the old package names.
* Consolidate some CI code.

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - David Gardner (https://github.com/dagardner-nv)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#181
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Feature: a tunable RAG evaluator

This PR adds a tunable RAG evaluator that allows for flexible evaluation of RAG workflows.
It includes a default scoring mechanism based on an expected answer description rather than a ground truth answer.

Key Points:
- Full control over judge LLM prompt
- Full control over scoring guidelines
- Everything is configurable from the config file
- No changes to dataset format other than answer format etc.

Usage Example:
```
eval:
  evaluators:
    custom_rag_evaluation:
      _type: tunable_rag_evaluator
      llm_name: nim_rag_eval_llm
      default_scoring: false
      default_score_weights:
        coverage: 0.5
        correctness: 0.3
        relevance: 0.2
      judge_llm_prompt: >
        You are an intelligent evaluator that scores the generated answer based on the description of the expected answer.
        The score is a measure of how well the generated answer matches the description of the expected answer based on the question.
        Take into account the question, the relevance of the answer to the question and the quality compared to the description of the expected answer.

        Rules:
        - The score must be a float of any value between 0.0 and 1.0 on a sliding scale.
        - The reasoning string must be concise and to the point. It should be 1 sentence and 2 only if extra description is needed. It must explain why the score was given and what is different between the generated answer and the expected answer.
```

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AgentIQ/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - https://github.com/liamy-nv
  - David Gardner (https://github.com/dagardner-nv)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)
  - Michael Demoret (https://github.com/mdemoret-nv)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#110
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
…DIA#184)

Closes NVIDIA#183

This PR fixes errors in the CLI documentation. The section that describes the CLI commands to update locally configured remote registry channels is incorrect.

The following command changes are implemented:

`aiq configure registry update` is modified to `aiq configure channel update`

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Matthew Penn (https://github.com/mpenn)

Approvers:
  - David Gardner (https://github.com/dagardner-nv)

URL: NVIDIA#184
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - David Gardner (https://github.com/dagardner-nv)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#187
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
* support python 3.11

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* fix doc

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update pyproject.toml

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update uv.lock

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update README to specify Python version requirements

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* resolve comments

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* fix comments

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* add support of ImageContext and AudioContext to AIQChatRequest

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

---------

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Rendered doc here:
https://github.com/AnuradhaKaruppiah/oss-agentiq/blob/doc-typo/examples/simple_calculator/README.md#accuracy-evaluation

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - David Gardner (https://github.com/dagardner-nv)

URL: NVIDIA#188
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Closes

This PR adds support tracing of AIQ intemediate steps and the underlying framework (Langchain, CrewAI, etc.) using Weights and Biases [Weave](https://weave-docs.wandb.ai) platform for better interactivity and off-the-shelf LLMOps life-cycle.

As a user all you need to do is update the `config.yaml` with the following:

```
general:
  use_uvloop: true
  telemetry:
    tracing:
      weave:
        _type: weave
        entity: <your-wandb-entity>
        project: my-awesome-genai-app
```

An example trace timeline logged to Weave:

<img width="1728" alt="image" src="https://github.com/user-attachments/assets/382d4926-b115-4514-a453-c58cf550a457" />

View more traces in this project: https://wandb.ai/ayut/simple_calculator/weave/traces?view=traces_default

cc: @mdemoret-nv

This PR was opened in favor of NVIDIA#135

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AgentIQ/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Ayush Thakur (https://github.com/ayulockin)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)

URL: NVIDIA#170
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Yuchen Zhang (https://github.com/yczhang-nv)

URL: NVIDIA#190
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
…IDIA#163)

* Note the previous name of the project to avoid confusing users.

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - David Gardner (https://github.com/dagardner-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#163
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
changed `concurrency` to `max_concurrency` in config parameters of _evaluate.md_

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Soumili Nandi (https://github.com/soumilinandi)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#191
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Closes

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Eric Evans II (https://github.com/ericevans-nv)

URL: NVIDIA#195
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
* support python 3.11

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* fix doc

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update pyproject.toml

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update uv.lock

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update README to specify Python version requirements

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* resolve comments

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* fix comments

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* updated simple example eval dataset

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

* update the nim_rag_eval_llm config

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>

---------

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
…ut.json (NVIDIA#198)

Closes

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)

URL: NVIDIA#198
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
…A#204)

Closes NVIDIA#202

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Eric Evans II (https://github.com/ericevans-nv)

URL: NVIDIA#204
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
This PR makes `opentelemetry` and `phoenix` as optional dependencies for Omniverse integration.

- Moved the telemetry dependencies from AIQ core dependency to a separate group of optional dependency, which can be installed by `uv pip install -e '.[telemetry]'`.
- Created `src/aiq/utils/optional_imports.py` to include optional import functionalities and dummy implementations. This module can be reused if we want to add more optional dependencies in the future.
- When the configuration file contains `tracing` section but the telemetry dependencies are not installed, `TelemetryOptionalImportError` will be raised
- Updated files that imports the optional dependencies to gracefully handle the case when those dependencies are not installed
- Updated logging and documentation

Closes [AIQ-1064](https://jirasw.nvidia.com/browse/AIQ-1064)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#167
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
…#130)

This pull request introduces the ability to extract and utilize user-defined attributes from incoming HTTP requests. These attributes can be accessed via the UserManagerAttributes class within the AIQContext and used in workflow tools.
## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AgentIQ/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Eric Evans II (https://github.com/ericevans-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#130
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Steps to repro:
Install latest AiqToolKit from source and run the simple-calculator example:
```
aiq run --config_file examples/simple_cal
culator/configs/config.yml --input "What is the product of 3 and 7, and is it greater than the curre
nt hour?"
```

Following traceback is seen:
```
2025-05-05 11:18:30,130 - aiq.cli.commands.start - ERROR - Failed to initialize workflow
Traceback (most recent call last):
  File "/home/devcontainers/dev/forks/agentiq/src/aiq/cli/commands/start.py", line 229, in invoke_subcommand
    return asyncio.run(run_plugin())
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/devcontainers/dev/forks/agentiq/.venv/lib/python3.12/site-packages/nest_asyncio.py", line 30, in run
    return loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/devcontainers/dev/forks/agentiq/.venv/lib/python3.12/site-packages/nest_asyncio.py", line 98, in run_until_complete
    return f.result()
           ^^^^^^^^^^
  File "/home/devcontainers/.local/share/uv/python/cpython-3.12.8-linux-x86_64-gnu/lib/python3.12/asyncio/futures.py", line 202, in result
    raise self._exception.with_traceback(self._exception_tb)
  File "/home/devcontainers/.local/share/uv/python/cpython-3.12.8-linux-x86_64-gnu/lib/python3.12/asyncio/tasks.py", line 316, in __step_run_and_handle_result
    result = coro.throw(exc)
             ^^^^^^^^^^^^^^^
  File "/home/devcontainers/dev/forks/agentiq/src/aiq/cli/commands/start.py", line 227, in run_plugin
    await front_end_plugin.run()
  File "/home/devcontainers/dev/forks/agentiq/src/aiq/front_ends/console/console_front_end_plugin.py", line 74, in run
    await self.run_workflow(session_manager)
  File "/home/devcontainers/dev/forks/agentiq/src/aiq/front_ends/console/console_front_end_plugin.py", line 94, in run_workflow
    runner_outputs = await asyncio.gather(*[run_single_query(query) for query in input_list])
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/devcontainers/.local/share/uv/python/cpython-3.12.8-linux-x86_64-gnu/lib/python3.12/asyncio/tasks.py", line 385, in __wakeup
    future.result()
  File "/home/devcontainers/.local/share/uv/python/cpython-3.12.8-linux-x86_64-gnu/lib/python3.12/asyncio/tasks.py", line 314, in __step_run_and_handle_result
    result = coro.send(None)
             ^^^^^^^^^^^^^^^
  File "/home/devcontainers/dev/forks/agentiq/src/aiq/front_ends/console/console_front_end_plugin.py", line 84, in run_single_query
    async with session_manager.session(user_input_callback=prompt_for_input_cli) as session:
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/devcontainers/.local/share/uv/python/cpython-3.12.8-linux-x86_64-gnu/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/devcontainers/dev/forks/agentiq/src/aiq/runtime/session.py", line 100, in session
    self.set_request_attributes(request)
  File "/home/devcontainers/dev/forks/agentiq/src/aiq/runtime/session.py", line 127, in set_request_attributes
    self._context.metadata._request.method = request.method
                                             ^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'method'
Error: 'NoneType' object has no attribute 'method'
```

This PR adds a sanity check to prevent request access unless it is present.

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Eric Evans II (https://github.com/ericevans-nv)

URL: NVIDIA#209
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
This PR adds a new multi-agent example for alert triaging.

Tested offline and verified the output to match expected labels on the test dataset (in the `data/` folder).

Authors:
  - https://github.com/hsin-c
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - David Gardner (https://github.com/dagardner-nv)

URL: NVIDIA#193
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Closes

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Eric Evans II (https://github.com/ericevans-nv)

Approvers:
  - David Gardner (https://github.com/dagardner-nv)

URL: NVIDIA#211
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
* All examples, plugins and transitional packages now declare a version dependency on `aiqtoolkit`
* Adds an `ci/release/update-version.sh` script to automate the updating of versions at release time.
* Add a `ci/release/update_toml_dep.py` (called by `ci/release/update-version.sh`) to parse and update individual `pyproject.toml` files.
* Add a new dev dependency on `tomlkit` (this was already present in the env as a transitive dep).
* Ensure that nightly builds declare explicit dependencies on the nightly version number

Closes NVIDIA#197

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - David Gardner (https://github.com/dagardner-nv)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#208
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
This PR introduces a new example agent, profiler agent.
Users can use this agent to chat with the profiling traces in their AgentIQ agent application and use it as an interactive profiling tool.

https://github.com/user-attachments/assets/318f9a1e-fb5a-47dd-a9f4-32e088aebd51

This agent requires users to configure AgentIQ with profiling and telemetry features enabled, following [the documentation](https://docs.nvidia.com/agentiq/latest/guides/profiler.html)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AgentIQ/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Zac Wang (https://github.com/zac-wang-nv)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#120
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
…example and fixed broken link in `README` (NVIDIA#210)

- Re-generated `uv.lock`
- Clean up `pyproject.toml` for profiler agent example and fix broken link in `README`

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)
  - Zac Wang (https://github.com/zac-wang-nv)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#210
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Closes NVIDIA#151

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Hritik Raj (https://github.com/Hritik003)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

Approvers:
  - David Gardner (https://github.com/dagardner-nv)

URL: NVIDIA#186
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Closes

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Victor Yudin (https://github.com/VictorYudin)

Approvers:
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)

URL: NVIDIA#215
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
yczhang-nv and others added 10 commits May 9, 2025 16:55
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
@mdemoret-nv mdemoret-nv requested a review from lvojtku May 29, 2025 00:34
@mdemoret-nv
Copy link
Collaborator

@lvojtku Need your review on the docs changes

Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
Signed-off-by: Yuchen Zhang <yuchenz@nvidia.com>
@yczhang-nv
Copy link
Contributor Author

@lvojtku I'll get this merged after having your approval on the documentation changes

@mdemoret-nv
Copy link
Collaborator

/merge

@rapids-bot rapids-bot bot merged commit 9e322f1 into NVIDIA:develop May 30, 2025
12 checks passed
@yczhang-nv yczhang-nv deleted the yuchen-fix-aiq-1213 branch May 30, 2025 16:57
gfreeman-nvidia pushed a commit to gfreeman-nvidia/AIQToolkit that referenced this pull request May 30, 2025
- Add support for AWS Bedrock LLM provider. The provider can be accessed by type `aws_bedrock`, and it is compatible with both `langchain` and `llamaindex`
- Added unit tests for existing combinations of existing LLM providers and LLM frameworks. The unit tests are skipped by default, since each of them needs credential to interact with LLM models, but it should be updated and checked each time we add/modify the LLM provider/client to make sure they are still working
- Fixed OpenAI + `llamaindex` LLM client that is not working

Closes [AIQ-1213](https://jirasw.nvidia.com/browse/AIQ-1213)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)
  - David Gardner (https://github.com/dagardner-nv)
  - https://github.com/liamy-nv
  - Matthew Penn (https://github.com/mpenn)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)
  - Ayush Thakur (https://github.com/ayulockin)
  - Soumili Nandi (https://github.com/soumilinandi)
  - Eric Evans II (https://github.com/ericevans-nv)
  - https://github.com/hsin-c
  - Zac Wang (https://github.com/zac-wang-nv)
  - Hritik Raj (https://github.com/Hritik003)
  - Victor Yudin (https://github.com/VictorYudin)
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)
  - Michael Demoret (https://github.com/mdemoret-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#238
Signed-off-by: Greg Freeman <gfreeman@nvidia.com>
ericevans-nv pushed a commit to ericevans-nv/agent-iq that referenced this pull request Jun 3, 2025
- Add support for AWS Bedrock LLM provider. The provider can be accessed by type `aws_bedrock`, and it is compatible with both `langchain` and `llamaindex`
- Added unit tests for existing combinations of existing LLM providers and LLM frameworks. The unit tests are skipped by default, since each of them needs credential to interact with LLM models, but it should be updated and checked each time we add/modify the LLM provider/client to make sure they are still working
- Fixed OpenAI + `llamaindex` LLM client that is not working

Closes [AIQ-1213](https://jirasw.nvidia.com/browse/AIQ-1213)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)
  - David Gardner (https://github.com/dagardner-nv)
  - https://github.com/liamy-nv
  - Matthew Penn (https://github.com/mpenn)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)
  - Ayush Thakur (https://github.com/ayulockin)
  - Soumili Nandi (https://github.com/soumilinandi)
  - Eric Evans II (https://github.com/ericevans-nv)
  - https://github.com/hsin-c
  - Zac Wang (https://github.com/zac-wang-nv)
  - Hritik Raj (https://github.com/Hritik003)
  - Victor Yudin (https://github.com/VictorYudin)
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)
  - Michael Demoret (https://github.com/mdemoret-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#238
Signed-off-by: Eric Evans <194135482+ericevans-nv@users.noreply.github.com>
ericevans-nv pushed a commit to ericevans-nv/agent-iq that referenced this pull request Jun 3, 2025
- Add support for AWS Bedrock LLM provider. The provider can be accessed by type `aws_bedrock`, and it is compatible with both `langchain` and `llamaindex`
- Added unit tests for existing combinations of existing LLM providers and LLM frameworks. The unit tests are skipped by default, since each of them needs credential to interact with LLM models, but it should be updated and checked each time we add/modify the LLM provider/client to make sure they are still working
- Fixed OpenAI + `llamaindex` LLM client that is not working

Closes [AIQ-1213](https://jirasw.nvidia.com/browse/AIQ-1213)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)
  - David Gardner (https://github.com/dagardner-nv)
  - https://github.com/liamy-nv
  - Matthew Penn (https://github.com/mpenn)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)
  - Ayush Thakur (https://github.com/ayulockin)
  - Soumili Nandi (https://github.com/soumilinandi)
  - Eric Evans II (https://github.com/ericevans-nv)
  - https://github.com/hsin-c
  - Zac Wang (https://github.com/zac-wang-nv)
  - Hritik Raj (https://github.com/Hritik003)
  - Victor Yudin (https://github.com/VictorYudin)
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)
  - Michael Demoret (https://github.com/mdemoret-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#238
Signed-off-by: Eric Evans <194135482+ericevans-nv@users.noreply.github.com>
ericevans-nv pushed a commit to ericevans-nv/agent-iq that referenced this pull request Jun 3, 2025
- Add support for AWS Bedrock LLM provider. The provider can be accessed by type `aws_bedrock`, and it is compatible with both `langchain` and `llamaindex`
- Added unit tests for existing combinations of existing LLM providers and LLM frameworks. The unit tests are skipped by default, since each of them needs credential to interact with LLM models, but it should be updated and checked each time we add/modify the LLM provider/client to make sure they are still working
- Fixed OpenAI + `llamaindex` LLM client that is not working

Closes [AIQ-1213](https://jirasw.nvidia.com/browse/AIQ-1213)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)
  - David Gardner (https://github.com/dagardner-nv)
  - https://github.com/liamy-nv
  - Matthew Penn (https://github.com/mpenn)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)
  - Ayush Thakur (https://github.com/ayulockin)
  - Soumili Nandi (https://github.com/soumilinandi)
  - Eric Evans II (https://github.com/ericevans-nv)
  - https://github.com/hsin-c
  - Zac Wang (https://github.com/zac-wang-nv)
  - Hritik Raj (https://github.com/Hritik003)
  - Victor Yudin (https://github.com/VictorYudin)
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)
  - Michael Demoret (https://github.com/mdemoret-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#238
Signed-off-by: Eric Evans <194135482+ericevans-nv@users.noreply.github.com>
AnuradhaKaruppiah pushed a commit to AnuradhaKaruppiah/oss-agentiq that referenced this pull request Aug 4, 2025
- Add support for AWS Bedrock LLM provider. The provider can be accessed by type `aws_bedrock`, and it is compatible with both `langchain` and `llamaindex`
- Added unit tests for existing combinations of existing LLM providers and LLM frameworks. The unit tests are skipped by default, since each of them needs credential to interact with LLM models, but it should be updated and checked each time we add/modify the LLM provider/client to make sure they are still working
- Fixed OpenAI + `llamaindex` LLM client that is not working

Closes [AIQ-1213](https://jirasw.nvidia.com/browse/AIQ-1213)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)
  - David Gardner (https://github.com/dagardner-nv)
  - https://github.com/liamy-nv
  - Matthew Penn (https://github.com/mpenn)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)
  - Ayush Thakur (https://github.com/ayulockin)
  - Soumili Nandi (https://github.com/soumilinandi)
  - Eric Evans II (https://github.com/ericevans-nv)
  - https://github.com/hsin-c
  - Zac Wang (https://github.com/zac-wang-nv)
  - Hritik Raj (https://github.com/Hritik003)
  - Victor Yudin (https://github.com/VictorYudin)
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)
  - Michael Demoret (https://github.com/mdemoret-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#238
scheckerNV pushed a commit to scheckerNV/aiq-factory-reset that referenced this pull request Aug 22, 2025
- Add support for AWS Bedrock LLM provider. The provider can be accessed by type `aws_bedrock`, and it is compatible with both `langchain` and `llamaindex`
- Added unit tests for existing combinations of existing LLM providers and LLM frameworks. The unit tests are skipped by default, since each of them needs credential to interact with LLM models, but it should be updated and checked each time we add/modify the LLM provider/client to make sure they are still working
- Fixed OpenAI + `llamaindex` LLM client that is not working

Closes [AIQ-1213](https://jirasw.nvidia.com/browse/AIQ-1213)

## By Submitting this PR I confirm:
- I am familiar with the [Contributing Guidelines](https://github.com/NVIDIA/AIQToolkit/blob/develop/docs/source/advanced/contributing.md).
- We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  - Any contribution which contains commits that are not Signed-Off will not be accepted.
- When the PR is ready for review, new or existing tests cover these changes.
- When the PR is ready for review, the documentation is up to date with these changes.

Authors:
  - Yuchen Zhang (https://github.com/yczhang-nv)
  - David Gardner (https://github.com/dagardner-nv)
  - https://github.com/liamy-nv
  - Matthew Penn (https://github.com/mpenn)
  - Anuradha Karuppiah (https://github.com/AnuradhaKaruppiah)
  - Ayush Thakur (https://github.com/ayulockin)
  - Soumili Nandi (https://github.com/soumilinandi)
  - Eric Evans II (https://github.com/ericevans-nv)
  - https://github.com/hsin-c
  - Zac Wang (https://github.com/zac-wang-nv)
  - Hritik Raj (https://github.com/Hritik003)
  - Victor Yudin (https://github.com/VictorYudin)
  - Dhruv Nandakumar (https://github.com/dnandakumar-nv)
  - Michael Demoret (https://github.com/mdemoret-nv)

Approvers:
  - Michael Demoret (https://github.com/mdemoret-nv)

URL: NVIDIA#238
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

improvement Improvement to existing functionality non-breaking Non-breaking change

Projects

None yet

Development

Successfully merging this pull request may close these issues.