Expose native GSPO through Tinker API

[kalectory]

• Expose SkyRL's existing native GSPO implementation through the Tinker-compatible API.
• Accept Tinker-style clipping thresholds and normalize them to SkyRL's epsilon configuration.

Future Work

• This path will fail on Jax, similar to the existing exposed DPPO and PPO_CRITIC paths. We generalize the PPO_CRITIC check to fail loudly on those paths too for now.

Testing

uv run --extra dev pre-commit run --files skyrl/backends/skyrl_train_backend.py skyrl/tinker/api.py skyrl/tinker/types.py

Verified formatting, lint, and secret checks on all changed files.

---

Note

Low Risk
Changes are API surface and backend routing for a new loss_fn name; JAX fails fast on unsupported losses. No auth or data-path changes; training behavior on SkyRL-Train depends on existing native GSPO implementation.

Overview
Adds gspo as a Tinker-compatible loss_fn on forward/backward, alongside existing PPO-style clip_low_threshold / clip_high_threshold config validation in the API layer.

The SkyRL-Train backend now treats gspo like ppo for threshold normalization (Tinker ratios → SkyRL epsilon fields) but maps the internal loss name to gspo instead of regular.

On JAX, unsupported losses are rejected with a single check against LOSS_TYPES (so gspo, dppo, and ppo_critic fail loudly). Tests cover that behavior via JaxBackendImpl._model_pass.

^{Reviewed by Cursor Bugbot for commit efb4101. Bugbot is set up for automated code reviews on this repo. Configure here.}

[gemini-code-assist]

Code Review

This pull request introduces support for the "gspo" loss function across the training backend and API. It updates the input validation schemas, config normalization logic, and adds corresponding unit tests to verify the new functionality. There are no review comments, so I have no feedback to provide.

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{Reviewed by Cursor Bugbot for commit 096639f. Configure here.}

[erictang000]

lgtm, ty!