Add litellm_proxy/gpt-oss-20b to resolve_model_config.py

[juanmichelini]

Summary

This PR adds the gpt-oss-20b model with id gpt-oss-20b to the expected models in resolve_model_config.py, as requested in issue #1905.

Changes:

• Added gpt-oss-20b model configuration to the MODELS dictionary with litellm_proxy/gpt-oss-20b as the model path
• Added gpt-oss-20b to the EXPECTED_MODELS list in the test file
• Added test_gpt_oss_20b_config test to verify the model configuration

Fixes #1905

Checklist

• If the PR is changing/adding functionality, are there tests to reflect this?
• If there is an example, have you run the example to make sure that it works?
• If there are instructions on how to run the code, have you followed the instructions and made sure that it works?
• If the feature is significant enough to require documentation, is there a PR open on the OpenHands/docs repository with the same branch name?
• Is the github CI passing?

@juanmichelini can click here to continue refining the PR

---

Agent Server images for this PR

• GHCR package: https://github.com/OpenHands/agent-sdk/pkgs/container/agent-server

Variants & Base Images

Variant	Architectures	Base Image	Docs / Tags
java	amd64, arm64	eclipse-temurin:17-jdk	Link
python	amd64, arm64	nikolaik/python-nodejs:python3.13-nodejs22	Link
golang	amd64, arm64	golang:1.21-bookworm	Link

Pull (multi-arch manifest)

# Each variant is a multi-arch manifest supporting both amd64 and arm64
docker pull ghcr.io/openhands/agent-server:45b772b-python

Run

docker run -it --rm \
  -p 8000:8000 \
  --name agent-server-45b772b-python \
  ghcr.io/openhands/agent-server:45b772b-python

All tags pushed for this build

ghcr.io/openhands/agent-server:45b772b-golang-amd64
ghcr.io/openhands/agent-server:45b772b-golang_tag_1.21-bookworm-amd64
ghcr.io/openhands/agent-server:45b772b-golang-arm64
ghcr.io/openhands/agent-server:45b772b-golang_tag_1.21-bookworm-arm64
ghcr.io/openhands/agent-server:45b772b-java-amd64
ghcr.io/openhands/agent-server:45b772b-eclipse-temurin_tag_17-jdk-amd64
ghcr.io/openhands/agent-server:45b772b-java-arm64
ghcr.io/openhands/agent-server:45b772b-eclipse-temurin_tag_17-jdk-arm64
ghcr.io/openhands/agent-server:45b772b-python-amd64
ghcr.io/openhands/agent-server:45b772b-nikolaik_s_python-nodejs_tag_python3.13-nodejs22-amd64
ghcr.io/openhands/agent-server:45b772b-python-arm64
ghcr.io/openhands/agent-server:45b772b-nikolaik_s_python-nodejs_tag_python3.13-nodejs22-arm64
ghcr.io/openhands/agent-server:45b772b-golang
ghcr.io/openhands/agent-server:45b772b-java
ghcr.io/openhands/agent-server:45b772b-python

About Multi-Architecture Support

• Each variant tag (e.g., 45b772b-python) is a multi-arch manifest supporting both amd64 and arm64
• Docker automatically pulls the correct architecture for your platform
• Individual architecture tags (e.g., 45b772b-python-amd64) are also available if needed

[all-hands-bot]

Overall, this PR looks good! The model configuration follows the existing patterns and the test coverage is appropriate. I have one minor suggestion about naming consistency.

[all-hands-bot]

🟡 Suggestion: Consider using "GPT-OSS-20B" for the display name to maintain consistency with other GPT models like "GPT-5.2" and "GPT-5.2 Codex" which use hyphens. However, if "GPT OSS 20B" is the official name, the current format is fine.

[openhands-ai]

Looks like there are a few issues preventing this PR from being merged!

• GitHub Actions are failing:
  • Run tests
• There are merge conflicts

If you'd like me to help, just leave a comment, like

@OpenHands please fix the merge conflicts on PR #1906 at branch `openhands/add-gpt-oss-20b-model`

or

@OpenHands please fix the failing actions on PR #1906 at branch `openhands/add-gpt-oss-20b-model`

Feel free to include any additional details that might help me get this PR into a better state.

_{^{You can manage your notification settings}}