Support on local deployment for SWE-Bench.

[zzfoutofspace]

Feature proposal

Do we have a plan on supporting local sandbox for SWE-Bench agentic training? Currently it seems only vefaas and modal are supported.

Motivation and use case

agent reinforcement learning

Related area

core agent loop

Proposed design or API

No response

Alternatives considered

No response

Additional context

No response

Contribution

• I am willing to help implement this feature.
• I am willing to help test or validate this feature.