[2026-11-03T14:47:17Z] GPU 0: Starting worker at PlasticBottlesInSquarePailTask/0 (resume=false) INFO:root:Codex source: local; executable: /data/NTU_slab/draven/proj/codex-main/codex-rs/target/debug/codex INFO:root:Harness action timeout: 1800 seconds; Codex stream idle timeout: 600 seconds INFO:root:Backend: codex; model: gpt-6-astra; effort: xhigh; decision log: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T144719.534865Z_eval.jsonl INFO:root:Episode statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T144719.534865Z_eval.episodes.jsonl INFO:root:Plan before execution: False INFO:root:Multi-action execution: True INFO:root:GPT action format: eef_absolute INFO:root:Manipulation notes included: True; selected file: note.md INFO:root:Adaptive action chunk: False INFO:root:Configured fixed action chunk size: 8 frames INFO:root:Async inference: False; reserved tail steps: 5 INFO:root:Async prompt included: False INFO:root:VLA as reference: False INFO:root:Shared memory: disabled; mode: text INFO:root:Serving harness on ws://127.0.0.1:8000 INFO:websockets.server:server listening on 127.0.0.1:8000 INFO:websockets.server:connection rejected (200 OK) INFO:websockets.server:connection open INFO:src.evaluation.robolab_bench.websocket_server:Connection from ('127.0.0.1', 33630) opened INFO:src.evaluation.robolab_bench.episode_stats:Episode 332fb8010ea14d20b3f7836ecd8531f0-0 (codex/gpt-6-astra) ended: wall=1044.688s harness=853.749s tokens=2807325 usage_complete=False decisions=67 executed_steps=529 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T144719.534865Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode 08d28bb3aff040cda2710597759c3a87-1 (codex/gpt-6-astra) ended: wall=1313.496s harness=1108.572s tokens=3797524 usage_complete=False decisions=71 executed_steps=561 steps_complete=True network_wait=19.651561804115772 rate_limit_wait=0 rate_limit_count=0 excluded=19.651561804115772 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T144719.534865Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode 4528300b5c4341fb85054ec16f6ba0f5-2 (codex/gpt-6-astra) execution_error: wall=2970.059s harness=2778.802s tokens=2888088 usage_complete=False decisions=68 executed_steps=544 steps_complete=False network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T144719.534865Z_eval.episodes.jsonl ERROR:websockets.server:connection handler failed Traceback (most recent call last): File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/site-packages/websockets/asyncio/server.py", line 376, in conn_handler await self.handler(connection) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 88, in _handler await self._serve_connection(websocket, managed=managed) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 124, in _serve_connection response = (await self._infer_harness(websocket, request) if managed File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 105, in _infer_harness return worker.result() File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/asyncio/threads.py", line 25, in to_thread return await loop.run_in_executor(None, func_call) File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/concurrent/futures/thread.py", line 58, in run result = self.fn(*self.args, **self.kwargs) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 149, in infer command = self._act(request) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 241, in _act return self.backend.act(request["observation"], control_step=request["control_step"]) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 471, in act return self._act_execute(observation, control_step=control_step, deadline=deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 600, in _act_execute event = self._next_event(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1122, in _next_event return self._queue.popleft() if self._queue else self._read(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1129, in _read raise TimeoutError("Timed out waiting for Codex") TimeoutError: Timed out waiting for Codex INFO:websockets.server:server closing INFO:websockets.server:server closed INFO:root:Harness server stopped [2026-11-03T16:19:26Z] GPU 0: PlasticBottlesInSquarePailTask/2: client exited (status=0) with incomplete results; retry 1/10 in 30s [2026-11-03T16:19:56Z] GPU 0: Starting worker at PlasticBottlesInSquarePailTask/2 (resume=--resume) INFO:root:Codex source: local; executable: /data/NTU_slab/draven/proj/codex-main/codex-rs/target/debug/codex INFO:root:Harness action timeout: 1800 seconds; Codex stream idle timeout: 600 seconds INFO:root:Backend: codex; model: gpt-6-astra; effort: xhigh; decision log: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T161958.984772Z_eval.jsonl INFO:root:Episode statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T161958.984772Z_eval.episodes.jsonl INFO:root:Plan before execution: False INFO:root:Multi-action execution: True INFO:root:GPT action format: eef_absolute INFO:root:Manipulation notes included: True; selected file: note.md INFO:root:Adaptive action chunk: False INFO:root:Configured fixed action chunk size: 8 frames INFO:root:Async inference: False; reserved tail steps: 5 INFO:root:Async prompt included: False INFO:root:VLA as reference: False INFO:root:Shared memory: disabled; mode: text INFO:root:Serving harness on ws://127.0.0.1:8000 INFO:websockets.server:server listening on 127.0.0.1:8000 INFO:websockets.server:connection rejected (200 OK) INFO:websockets.server:connection open INFO:src.evaluation.robolab_bench.websocket_server:Connection from ('127.0.0.1', 39676) opened INFO:src.evaluation.robolab_bench.episode_stats:Episode e29aea49ac2546fba7b0302df09e68c1-2 (codex/gpt-6-astra) execution_error: wall=3000.690s harness=2782.891s tokens=2416259 usage_complete=False decisions=75 executed_steps=600 steps_complete=False network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T161958.984772Z_eval.episodes.jsonl ERROR:websockets.server:connection handler failed Traceback (most recent call last): File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/site-packages/websockets/asyncio/server.py", line 376, in conn_handler await self.handler(connection) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 88, in _handler await self._serve_connection(websocket, managed=managed) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 124, in _serve_connection response = (await self._infer_harness(websocket, request) if managed File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 105, in _infer_harness return worker.result() File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/asyncio/threads.py", line 25, in to_thread return await loop.run_in_executor(None, func_call) File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/concurrent/futures/thread.py", line 58, in run result = self.fn(*self.args, **self.kwargs) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 149, in infer command = self._act(request) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 241, in _act return self.backend.act(request["observation"], control_step=request["control_step"]) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 471, in act return self._act_execute(observation, control_step=control_step, deadline=deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 600, in _act_execute event = self._next_event(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1122, in _next_event return self._queue.popleft() if self._queue else self._read(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1129, in _read raise TimeoutError("Timed out waiting for Codex") TimeoutError: Timed out waiting for Codex INFO:websockets.server:server closing INFO:websockets.server:server closed INFO:root:Harness server stopped [2026-11-03T17:11:31Z] GPU 0: PlasticBottlesInSquarePailTask/2: client exited (status=0) with incomplete results; retry 2/10 in 60s [2026-11-03T17:12:31Z] GPU 0: Starting worker at PlasticBottlesInSquarePailTask/2 (resume=--resume) INFO:root:Codex source: local; executable: /data/NTU_slab/draven/proj/codex-main/codex-rs/target/debug/codex INFO:root:Harness action timeout: 1800 seconds; Codex stream idle timeout: 600 seconds INFO:root:Backend: codex; model: gpt-6-astra; effort: xhigh; decision log: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T171234.289986Z_eval.jsonl INFO:root:Episode statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T171234.289986Z_eval.episodes.jsonl INFO:root:Plan before execution: False INFO:root:Multi-action execution: True INFO:root:GPT action format: eef_absolute INFO:root:Manipulation notes included: True; selected file: note.md INFO:root:Adaptive action chunk: False INFO:root:Configured fixed action chunk size: 8 frames INFO:root:Async inference: False; reserved tail steps: 5 INFO:root:Async prompt included: False INFO:root:VLA as reference: False INFO:root:Shared memory: disabled; mode: text INFO:root:Serving harness on ws://127.0.0.1:8000 INFO:websockets.server:server listening on 127.0.0.1:8000 INFO:websockets.server:connection rejected (200 OK) INFO:websockets.server:connection open INFO:src.evaluation.robolab_bench.websocket_server:Connection from ('127.0.0.1', 48210) opened INFO:src.evaluation.robolab_bench.episode_stats:Episode 95fb820f78d046ac8c42a12d580a1a23-2 (codex/gpt-6-astra) execution_error: wall=4803.296s harness=4273.755s tokens=7704209 usage_complete=False decisions=187 executed_steps=1496 steps_complete=False network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T171234.289986Z_eval.episodes.jsonl ERROR:websockets.server:connection handler failed Traceback (most recent call last): File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/site-packages/websockets/asyncio/server.py", line 376, in conn_handler await self.handler(connection) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 88, in _handler await self._serve_connection(websocket, managed=managed) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 124, in _serve_connection response = (await self._infer_harness(websocket, request) if managed File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 105, in _infer_harness return worker.result() File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/asyncio/threads.py", line 25, in to_thread return await loop.run_in_executor(None, func_call) File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/concurrent/futures/thread.py", line 58, in run result = self.fn(*self.args, **self.kwargs) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 149, in infer command = self._act(request) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 241, in _act return self.backend.act(request["observation"], control_step=request["control_step"]) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 471, in act return self._act_execute(observation, control_step=control_step, deadline=deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 600, in _act_execute event = self._next_event(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1122, in _next_event return self._queue.popleft() if self._queue else self._read(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1129, in _read raise TimeoutError("Timed out waiting for Codex") TimeoutError: Timed out waiting for Codex INFO:websockets.server:server closing INFO:websockets.server:server closed INFO:root:Harness server stopped [2026-11-03T18:34:11Z] GPU 0: PlasticBottlesInSquarePailTask/2: client exited (status=0) with incomplete results; retry 3/10 in 120s [2026-11-03T18:36:11Z] GPU 0: Starting worker at PlasticBottlesInSquarePailTask/2 (resume=--resume) INFO:root:Codex source: local; executable: /data/NTU_slab/draven/proj/codex-main/codex-rs/target/debug/codex INFO:root:Harness action timeout: 1800 seconds; Codex stream idle timeout: 600 seconds INFO:root:Backend: codex; model: gpt-6-astra; effort: xhigh; decision log: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T183614.673624Z_eval.jsonl INFO:root:Episode statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T183614.673624Z_eval.episodes.jsonl INFO:root:Plan before execution: False INFO:root:Multi-action execution: True INFO:root:GPT action format: eef_absolute INFO:root:Manipulation notes included: True; selected file: note.md INFO:root:Adaptive action chunk: False INFO:root:Configured fixed action chunk size: 8 frames INFO:root:Async inference: False; reserved tail steps: 5 INFO:root:Async prompt included: False INFO:root:VLA as reference: False INFO:root:Shared memory: disabled; mode: text INFO:root:Serving harness on ws://127.0.0.1:8000 INFO:websockets.server:server listening on 127.0.0.1:8000 INFO:websockets.server:connection rejected (200 OK) INFO:websockets.server:connection open INFO:src.evaluation.robolab_bench.websocket_server:Connection from ('127.0.0.1', 44534) opened INFO:src.evaluation.robolab_bench.episode_stats:Episode 780f21e0e7834ab9ade19e63433f237d-2 (codex/gpt-6-astra) execution_error: wall=3347.450s harness=3088.506s tokens=4223994 usage_complete=False decisions=92 executed_steps=736 steps_complete=False network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T183614.673624Z_eval.episodes.jsonl ERROR:websockets.server:connection handler failed Traceback (most recent call last): File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/site-packages/websockets/asyncio/server.py", line 376, in conn_handler await self.handler(connection) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 88, in _handler await self._serve_connection(websocket, managed=managed) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 124, in _serve_connection response = (await self._infer_harness(websocket, request) if managed File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/websocket_server.py", line 105, in _infer_harness return worker.result() File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/asyncio/threads.py", line 25, in to_thread return await loop.run_in_executor(None, func_call) File "/data/NTU_slab/draven/anaconda3/envs/codebase/lib/python3.10/concurrent/futures/thread.py", line 58, in run result = self.fn(*self.args, **self.kwargs) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 149, in infer command = self._act(request) File "/data/NTU_slab/draven/proj/codebase/src/evaluation/robolab_bench/harness_policy.py", line 241, in _act return self.backend.act(request["observation"], control_step=request["control_step"]) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 471, in act return self._act_execute(observation, control_step=control_step, deadline=deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 600, in _act_execute event = self._next_event(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1122, in _next_event return self._queue.popleft() if self._queue else self._read(deadline) File "/data/NTU_slab/draven/proj/codebase/src/agent/client.py", line 1129, in _read raise TimeoutError("Timed out waiting for Codex") TimeoutError: Timed out waiting for Codex INFO:websockets.server:server closing INFO:websockets.server:server closed INFO:root:Harness server stopped [2026-11-03T19:33:31Z] GPU 0: PlasticBottlesInSquarePailTask/2: client exited (status=0) with incomplete results; retry 4/10 in 240s [2026-11-03T19:37:31Z] GPU 0: Starting worker at PlasticBottlesInSquarePailTask/2 (resume=--resume) INFO:root:Codex source: local; executable: /data/NTU_slab/draven/proj/codex-main/codex-rs/target/debug/codex INFO:root:Harness action timeout: 1800 seconds; Codex stream idle timeout: 600 seconds INFO:root:Backend: codex; model: gpt-6-astra; effort: xhigh; decision log: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.jsonl INFO:root:Episode statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:root:Plan before execution: False INFO:root:Multi-action execution: True INFO:root:GPT action format: eef_absolute INFO:root:Manipulation notes included: True; selected file: note.md INFO:root:Adaptive action chunk: False INFO:root:Configured fixed action chunk size: 8 frames INFO:root:Async inference: False; reserved tail steps: 5 INFO:root:Async prompt included: False INFO:root:VLA as reference: False INFO:root:Shared memory: disabled; mode: text INFO:root:Serving harness on ws://127.0.0.1:8000 INFO:websockets.server:server listening on 127.0.0.1:8000 INFO:websockets.server:connection rejected (200 OK) INFO:websockets.server:connection open INFO:src.evaluation.robolab_bench.websocket_server:Connection from ('127.0.0.1', 52198) opened INFO:src.evaluation.robolab_bench.episode_stats:Episode 738e0efa864742e1aa6d18dd81c91736-2 (codex/gpt-6-astra) ended: wall=722.526s harness=599.600s tokens=2178263 usage_complete=False decisions=43 executed_steps=337 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode cbc51ba782744c1a859a401c532237d3-0 (codex/gpt-6-astra) ended: wall=600.525s harness=503.573s tokens=1887192 usage_complete=False decisions=35 executed_steps=273 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode 31d3266fd6354940bb9a2a5c4b2a5e47-1 (codex/gpt-6-astra) ended: wall=622.368s harness=543.033s tokens=2214624 usage_complete=True decisions=29 executed_steps=225 steps_complete=True network_wait=45.785648711025715 rate_limit_wait=0 rate_limit_count=0 excluded=45.785648711025715 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode c4b3db38d4da4dde820cc8f92a851ac5-2 (codex/gpt-6-astra) ended: wall=562.201s harness=499.221s tokens=1592304 usage_complete=True decisions=22 executed_steps=173 steps_complete=True network_wait=4.930022820830345 rate_limit_wait=0 rate_limit_count=0 excluded=4.930022820830345 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode 0e0e3448bcd7409b94b5932aa1375fd7-0 (codex/gpt-6-astra) ended: wall=985.752s harness=839.900s tokens=3181857 usage_complete=False decisions=60 executed_steps=473 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode e7ea93b494f34c5c9df14367693a6329-1 (codex/gpt-6-astra) ended: wall=550.681s harness=496.569s tokens=1323152 usage_complete=False decisions=22 executed_steps=172 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode 12e43cdfed3b4580aad3bf91c5873620-2 (codex/gpt-6-astra) ended: wall=295.113s harness=262.468s tokens=1055871 usage_complete=False decisions=13 executed_steps=103 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode b72beeb966794005ab1aff1ab8b0a72a-0 (codex/gpt-6-astra) ended: wall=445.608s harness=410.779s tokens=770979 usage_complete=False decisions=18 executed_steps=142 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode 3f35b21d2ffd471090b2310483a6d470-1 (codex/gpt-6-astra) ended: wall=460.925s harness=429.128s tokens=1459633 usage_complete=True decisions=16 executed_steps=127 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.episode_stats:Episode 24216372a3144b2d884a8f6f28d1a14f-2 (codex/gpt-6-astra) ended: wall=1012.832s harness=941.698s tokens=3362687 usage_complete=False decisions=38 executed_steps=300 steps_complete=True network_wait=0 rate_limit_wait=0 rate_limit_count=0 excluded=0.0 timing=complete; statistics: /data/NTU_slab/draven/proj/codebase/outputs/robolab_exploration/embo_multiact/gpu_0/harness/20261103T193734.435264Z_eval.episodes.jsonl INFO:src.evaluation.robolab_bench.websocket_server:Connection from ('127.0.0.1', 52198) closed INFO:websockets.server:server closing INFO:websockets.server:server closed INFO:root:Harness server stopped