Improve the UX for Adding Models in AutoGen Studio

[thistleknot]

What happened?

Describe the bug
A clear and concise description of what the bug is.
If it is a question or suggestion, please use Discussions
instead.

To Reproduce
Steps to reproduce the behavior. Please include code and outputs such as stacktrace.

python 3.10
pip venv
windows 11

(autogen) C:\Users\user\autogen>pip list installed
Package Version

---

aiofiles 24.1.0
alembic 1.15.2
annotated-types 0.7.0
anthropic 0.51.0
anyio 4.9.0
argon2-cffi 23.1.0
argon2-cffi-bindings 21.2.0
arrow 1.3.0
asttokens 3.0.0
async-lru 2.0.5
asyncer 0.0.8
attrs 25.3.0
autogen-agentchat 0.4.9.3
autogen-core 0.4.9.3
autogen-ext 0.4.9.3
autogenstudio 0.4.2.1
azure-ai-documentintelligence 1.0.2
azure-ai-inference 1.0.0b9
azure-core 1.34.0
azure-identity 1.22.0
babel 2.17.0
beautifulsoup4 4.13.4
bleach 6.2.0
certifi 2025.4.26
cffi 1.17.1
charset-normalizer 3.4.2
click 8.1.8
cobble 0.1.4
colorama 0.4.6
coloredlogs 15.0.1
comm 0.2.2
contourpy 1.3.2
cryptography 44.0.3
cycler 0.12.1
debugpy 1.8.14
decorator 5.2.1
defusedxml 0.7.1
Deprecated 1.2.18
diskcache 5.6.3
distro 1.9.0
dnspython 2.7.0
docker 7.1.0
email_validator 2.2.0
et_xmlfile 2.0.0
eval_type_backport 0.2.2
exceptiongroup 1.2.2
executing 2.2.0
fastapi 0.115.12
fastapi-cli 0.0.7
fastjsonschema 2.21.1
flatbuffers 25.2.10
fonttools 4.57.0
fqdn 1.5.1
greenlet 3.2.1
h11 0.16.0
html2text 2025.4.15
httpcore 1.0.9
httptools 0.6.4
httpx 0.28.1
humanfriendly 10.0
idna 3.10
importlib_metadata 8.6.1
ipykernel 6.29.5
ipython 8.36.0
isodate 0.7.2
isoduration 20.11.0
jedi 0.19.2
Jinja2 3.1.6
jiter 0.9.0
json5 0.12.0
jsonpointer 3.0.0
jsonref 1.1.0
jsonschema 4.23.0
jsonschema-specifications 2025.4.1
jupyter_client 8.6.3
jupyter_core 5.7.2
jupyter-events 0.12.0
jupyter-lsp 2.2.5
jupyter_server 2.15.0
jupyter_server_terminals 0.5.3
jupyterlab 4.4.2
jupyterlab_pygments 0.3.0
jupyterlab_server 2.27.3
kiwisolver 1.4.8
loguru 0.7.3
lxml 5.4.0
Mako 1.3.10
mammoth 1.9.0
markdown-it-py 3.0.0
markdownify 1.1.0
markitdown 0.0.2
MarkupSafe 3.0.2
matplotlib-inline 0.1.7
mdurl 0.1.2
mistralai 1.7.0
mistune 3.1.3
mpmath 1.3.0
msal 1.32.3
msal-extensions 1.3.1
nbclient 0.10.2
nbconvert 7.16.6
nbformat 5.10.4
nest-asyncio 1.6.0
notebook_shim 0.2.4
numpy 2.2.5
olefile 0.47
ollama 0.4.8
openai 1.77.0
openpyxl 3.1.5
opentelemetry-api 1.32.1
overrides 7.7.0
packaging 25.0
pandas 2.2.3
pandocfilters 1.5.1
parso 0.8.4
pathvalidate 3.2.3
pdfminer.six 20250506
pillow 11.2.1
pip 25.1.1
platformdirs 4.3.8
playwright 1.52.0
prometheus_client 0.21.1
prompt_toolkit 3.0.51
protobuf 5.29.4
psutil 7.0.0
psycopg 3.2.7
pure_eval 0.2.3
puremagic 1.29
pyautogen 0.9.0
pycparser 2.22
pydantic 2.11.4
pydantic_core 2.33.2
pydantic-settings 2.9.1
pydub 0.25.1
pyee 13.0.0
Pygments 2.19.1
PyJWT 2.10.1
pyparsing 3.2.3
pyreadline3 3.5.4
python-dateutil 2.9.0.post0
python-dotenv 1.1.0
python-json-logger 3.3.0
python-multipart 0.0.20
python-pptx 1.0.2
pytz 2025.2
pywin32 310
pywinpty 2.0.15
PyYAML 6.0.2
pyzmq 26.4.0
referencing 0.36.2
regex 2024.11.6
requests 2.32.3
rfc3339-validator 0.1.4
rfc3986-validator 0.1.1
rich 14.0.0
rich-toolkit 0.14.5
rpds-py 0.24.0
Send2Trash 1.8.3
setuptools 57.4.0
shellingham 1.5.4
six 1.17.0
sniffio 1.3.1
soupsieve 2.7
SpeechRecognition 3.14.2
SQLAlchemy 2.0.40
sqlmodel 0.0.24
stack-data 0.6.3
starlette 0.46.2
sympy 1.14.0
termcolor 3.1.0
terminado 0.18.1
tiktoken 0.9.0
tinycss2 1.4.0
tomli 2.2.1
tornado 6.4.2
tqdm 4.67.1
traitlets 5.14.3
typer 0.15.3
types-python-dateutil 2.9.0.20241206
typing_extensions 4.13.2
typing-inspection 0.4.0
tzdata 2025.2
uri-template 1.3.0
urllib3 2.4.0
uvicorn 0.34.2
watchfiles 1.0.5
wcwidth 0.2.13
webcolors 24.11.1
webencodings 0.5.1
websocket-client 1.8.0
websockets 15.0.1
win32_setctime 1.2.0
wrapt 1.17.2
xlrd 2.0.1
XlsxWriter 3.2.3
youtube-transcript-api 1.0.3
zipp 3.21.0

• If your input is just "I tried X, and it didn't work" or
  "X is not working", your issue will be ignored.
• If your input is not well formatted, it will hurt readability and
  may be ignored as well.

if I set the default_model using json_schema to

{
"provider": "autogen_ext.models.ollama.OllamaChatCompletionClient",
"component_type": "model",
"version": 1,
"component_version": 1,
"description": "Chat completion client for Ollama hosted models.",
"label": "OllamaChatCompletionClient",
"config": {
"model": "ertghiu256/qwen3-4b-code-reasoning",
"model_info": {
"family": "unknown",
"model_name": "ertghiu256/qwen3-4b-code-reasoning",
"vision": false,
"function_calling": true,
"json_output": false,
"structured_output": true
},
"base_url": "http://127.0.0.1:11434"
}
}

I can get the model to test successfully, but upon saving, the settings don't stick and the default open_ai model returns

Expected behavior

saving the default_model to a new schema should take effect upon reopening the default_model edit menu and drilling down into json_schema

Screenshots

step 1: save and test successfully, exit default model menu

step 2: after reopening default model

Additional context
Add any other context about the problem here.

Which packages was the bug in?

AutoGen Studio (autogensudio)

AutoGen library version.

Python dev (main branch)

Other library version.

No response

Model used

No response

Model provider

None

Other model provider

No response

Python version

None

.NET version

None

Operating system

None

[emanuilo]

same

[victordibia]

Thanks for the issue @thistleknot ,
Added to the backlog.
Btw, default model is meant for future internal use by the app (e.g., generating readable titles for sessions etc), what is your main/expected use case for setting the default model?

[thistleknot]

I'm trying to use autogen's capability with other provider's. I'm reading AI Agents in Action Chapter 4 on AutoGen, and I wanted to use OSS models, either openrouter, ollama, or mistral. There is an open case with mistral and extra parms returned in the response, so I fell back to ollama. Maybe I'm missing something, but how does one configure alternative models for use within autogen? Am I looking in the wrong place?

…

On Sat, May 10, 2025 at 10:15 AM Victor Dibia ***@***.***> wrote: *victordibia* left a comment (microsoft/autogen#6500) <#6500 (comment)> Thanks for the issue @thistleknot <https://github.com/thistleknot> , Added to the backlog. Btw, default model is meant for internal use by the app, what is your main use case for setting the default model? — Reply to this email directly, view it on GitHub <#6500 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABHKKOXS6JOPOIUFN5YSXZL25YXZXAVCNFSM6AAAAAB42MRSL2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQNRZGAZTMMJSHA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[victordibia]

Hi @thistleknot ,
Please see a tutorial on this here https://youtu.be/oum6EI7wohM?si=-c5mxrprMDgqIn2r&t=1050
Does that help?

See this video for a walkthrough of features - https://youtu.be/oum6EI7wohM

[thistleknot]

great that helped, although not necessarily the right place, but I now see how to add models gallery -> components -> add model

…

On Sat, May 10, 2025 at 10:27 AM Victor Dibia ***@***.***> wrote: *victordibia* left a comment (microsoft/autogen#6500) <#6500 (comment)> Hi @thistleknot <https://github.com/thistleknot> , Please see a tutorial on this here https://youtu.be/oum6EI7wohM?si=-c5mxrprMDgqIn2r&t=1050 Does that help? See this video for a walkthrough of features - https://youtu.be/oum6EI7wohM [image: Designing With AI] <https://www.youtube.com/watch?v=oum6EI7wohM> — Reply to this email directly, view it on GitHub <#6500 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABHKKOSGMJ34G44ZXMKVDW325YZIXAVCNFSM6AAAAAB42MRSL2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQNRZGA2DCOBWGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[victordibia]

Thanks for the feedback. I agree that the UX for adding new models could be improved.
I updated the title of the issue and we can use it to track updates and improvements here