microsoft / microsoft/semantic-kernel
.Net: Python: New Feature: Add `thinking_level` to Gemini connector
Nobody has claimed this yet.
- Dominant language
- C#
- Stars
- 28.6k
- Forks
- 4.8k
- Avg merge
- 14h 13m
- Merged PRs (30d)
- 18
Description
name: Add thinking_level to Gemini Python connector
about: Allow users to specify thinking levels
Gemini 3+ models support specifying thinking_level, which gives users control over their token and latency budgets.
Older models support a thinking_budget instead, but for simplicity I am proposing only support for the latest feature.
The .NET library seems to support this already: https://github.com/microsoft/semantic-kernel/blob/2a719ca3182f547203f54257f69c47a175fbb3fd/dotnet/src/Connectors/Connectors.Google/GeminiThinkingConfig.cs#L30-L32
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating the Python Gemini connector and compare its thinking configuration with the referenced .NET file, GeminiThinkingConfig.cs. Add support for the documented thinking_level option for Gemini 3+ models, then run the connector's existing tests and verify that users can pass the setting through to Gemini.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai, api
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 55/100