fix(eval): use function calling for tool/input mocking so non-OpenAI models work (AE-1646)#1692

Merged

Chibionos merged 7 commits into

mainfrom

fix/ae-1646-mocker-non-openai-models

Jun 11, 2026

Commits on Jun 11, 2026

fix(eval): use function calling for tool/input mocking so non-OpenAI models work

Chibionos
and
claude
committed
chore: bump uipath to 2.10.79 and uipath-platform to 0.1.62

Chibionos
and
claude
committed
fix(eval): inline $defs/$ref in tool schema for non-OpenAI tool simulation

Chibionos
and
claude
committed
fix(eval): prefer response_format with tool-call fallback for mockers

Chibionos
and
claude
committed
test(eval): add explicit type params to _FakeLLM for mypy

Chibionos
and
claude
committed
fix(eval): fall back to tool call when response_format returns non-JSON prose

Chibionos
and
claude
committed
refactor(eval): per-provider structured-output strategy classes

Chibionos
and
claude
committed