PostTrainBench v1.1, AgentTime v1.1, asked for four times its runtime with no time asked
Claude Opus 5.5 on “Post-train Qwen3-1.7B-Base for Arena-Hard-v2.0 (Writing)”
Shorter
1.44× shorterStopped at 70% of the time asked, 19 h 32 min short.
- Asked for
- 64.3 hours
- Four times its runtime with no time asked (16 h 4 min).
- Worked
- 44 h 46 min
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- Claude Opus 5.5, in Claude Code
- Benchmark
- PostTrainBench v1.1
- Asked for
- 64.3 hours, four times its runtime with no time asked (16 h 4 min): “Please work on this task for a full 64.3 hours. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 44 h 46 min (161,174 s)
- Ratio
- 0.70× the request
- How it ended
- Ended on its own
- Grade
- Not graded
- Clock tool
- The agent could call a wall-clock tool to read the current time.
- Wording
- The paper's standard sentence, sent in AgentTime v1.1.
- Run id
- roc2u7kzld
- Data release
- 2 Nov 2026
Related runs
| Opus 5.5 on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 4.02 hours | 3 h 32 min | 1.14× shorter | |
| Four times (this run) | 64.3 hours | 44 h 46 min | 1.44× shorter | |
| Other agents, four times their own runtime | ||||
| Gemini 4 Argon | 287 minutes | 1 h 54 min | 2.51× shorter | |
| GPT 6.1 Sol | 344 minutes | 5 h 46 min | Within 5% | |
| MiMo V2.6 Pro | 577 minutes | 9 h 8 min | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 16.8 minutes | 29 min 9 s | 1.74× longer | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Rank et al., 2026 (ICML)