Skip to content

Update Extensions for Harmony/GPT-OSS Tool Calling Support #2200

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
kunal-vaishnavi merged 7 commits into from
Jun 10, 2026
Merged

Update Extensions for Harmony/GPT-OSS Tool Calling Support #2200

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
3c75eed
Support Harmony/GPT-OSS tool calling: update ort-extensions, add tool…
May 12, 2026
85d265a
Merge main, update ort-extensions to merged Harmony PR (9780925)
Jun 8, 2026
2f149ba
Merge branch 'main' of https://github.com/microsoft/onnxruntime-genai…
Jun 9, 2026
58f64ec
update to newer ext commit with perf improvements
Jun 9, 2026
00c1f9c
undo model-qa.py
Jun 9, 2026
b0149a4
undo model-qa.py
Jun 9, 2026
ba3f00f
amend
Jun 9, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion cmake/deps.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ pybind11;https://github.com/pybind/pybind11/archive/refs/tags/v2.13.6.zip;f78029
googletest;https://github.com/google/googletest/archive/530d5c8c84abd2a46f38583ee817743c9b3a42b4.zip;5e3a61db2aa975cfd0f97ba92c818744e7fa7034
microsoft_wil;https://github.com/microsoft/wil/archive/refs/tags/v1.0.230629.1.zip;e4a542a323c070376f7c2d1973d0f7ddbc1d2fa5
directx_headers;https://github.com/microsoft/DirectX-Headers/archive/refs/tags/v1.613.1.zip;47653509a3371eabb156360f42faf582f314bf2e
onnxruntime_extensions;https://github.com/microsoft/onnxruntime-extensions.git;b077a8350da914fcb4a7e2c528e6659e67087da8
onnxruntime_extensions;https://github.com/microsoft/onnxruntime-extensions.git;97809255e4cfc6fdee2b32d565935067c0309f8b

# These two dependencies are for the optional constrained decoding feature (USE_GUIDANCE)
llguidance;https://github.com/microsoft/llguidance.git;94fa39128ef184ffeda33845f6d333f332a34b4d
Expand Down
7 changes: 7 additions & 0 deletions examples/python/model-qa.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,6 +61,13 @@ def main(args):
tool_call_end=args.tool_call_end,
)
input_list[0]["tools"] = tools
elif args.tools_file:
# Tools provided without constrained decoding (e.g. Harmony/GPT-OSS models
# that handle tool calling natively via their chat template).
# Tools are passed as a separate parameter to apply_chat_template,
# NOT embedded in messages (that's the Phi-4 constrained decoding path).
with open(args.tools_file, 'r') as f:
tools = json.dumps(json.load(f))

# Keep track of timings if requested
if args.timings:
Expand Down
Loading