Search-SWE
CPU TASK Valid result

Narrative-Clue Search

GPT-6 Luna · Codex (v0.164.0) · reasoning effort max

Task score (%)6.2
Input Tokens39,224,828including 38,358,508 cached
Output Tokens321,145
Agent time2h 0min
Cost$1.05
Recorded steps149

Cost is estimated from recorded agent tokens at official API rates in USD (2026-11-01; DeepSeek off-peak pricing).

Loading trajectory…

Select this tab to browse the saved files.

Coding time & token usage
Original agent completion
agent_timeout
Coding started
2026-10-28T23:04:14.708076Z
Coding finished
2026-10-29T01:04:17.069294Z
Input (includes cache)
39,224,828
Cached input (subset)
38,358,508
Cache writes (subset)
783,230
Output
321,145
Reasoning (reported separately; not added)
198,245
Recorded responses
144

Input includes cached input; output is counted once. Auxiliary reasoning counts are shown as recorded. Regrading does not repeat coding usage.

Select this tab to browse the saved files.

Configuration

Evaluation settings summary
{
  "agent": "Codex",
  "agent_timeout_sec": null,
  "coding_commit": null,
  "effort": "max",
  "model": "openai/gpt-6-luna",
  "provider": "OpenRouter",
  "resources": null,
  "run_mode": "official_harbor_regrade",
  "task_commit": "1d73c6f9c4a90ba922b7572d9395a6f40661bbb2",
  "verifier_commit": "1d73c6f9c4a90ba922b7572d9395a6f40661bbb2",
  "version": "0.164.0"
}

Select this tab to browse the saved files.

Resolved configuration

Select this tab to browse the saved files.

Trial log

Select this tab to browse the saved files.

Exceptions

Select this tab to browse the saved files.

Select this tab to browse the saved files.

Trajectory downloads & provenance

Public text redacts private values and normalizes host paths. Source hashes identify the original records.

Trial ID
narrative-clue-search-gpt-6-luna-max-1cf5c6415fc1
Source trajectory SHA-256
ac3195d0ea3530c3df536b71f434d3278a71d1d00330a802019a4b1c3ce821bc
Final reward SHA-256
06f7aec8653cc276261f7847c7bf07f8e91590faa8dc08e17c528d2464454c91