mirror of
https://github.com/agentscope-ai/ReMe.git
synced 2026-10-23 05:35:30 +00:00
* Add Docker benchmark sandbox and pooled runner * Fix benchmark worker reuse after Docker command timeout * Simplify benchmark worker requests and preserve artifacts * Harden benchmark sandbox reuse and isolate query state * Deduplicate benchmark sandbox cleanup and artifact checks * Report benchmark cleanup failures and preserve diagnostic artifacts * Handle optional Docker dependencies in benchmark unit tests * Keep benchmark Docker dependencies optional in CI * Fix cancellation of thread-pool background jobs * Finish background worker cleanup before propagating cancellation * Revert background worker cancellation changes * Fix benchmark worker plugin defaults before config validation * Stop interrupted benchmark containers before collecting diagnostics * Fix candidate exclusions and benchmark upload serialization * Fix late Docker creation cleanup and preserve query artifacts * Stream benchmark workspace archives and extract plan validation --------- Co-authored-by: jinli.yl <jinli.yl@alibaba-inc.com> |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| _workspace_fixture.py | ||
| run_reme_e2e_local.py | ||
| test_agent_session.py | ||
| test_auto_dream.py | ||
| test_auto_memory.py | ||
| test_auto_resource.py | ||
| test_auto_tag.py | ||
| test_benchmark_docker.py | ||
| test_codex_agent_wrapper.py | ||
| test_codex_plugin_install.py | ||
| test_codex_plugin_status_ui.py | ||
| test_embedding.py | ||
| test_llm.py | ||
| test_reme_e2e.py | ||
| test_stream_llm.py | ||