AppWorld, AgentTime v1.1, asked for a quarter of its own forecast
Claude Opus 5.5 on “Build a recruiting CSV and archive each candidate’s attachments”
Shorter
1.50× shorterStopped at 67% of the time asked, 55 s short.
- Asked for
- 2.75 minutes
- A quarter of its own forecast of 11 min.
- Worked
- 1 min 50 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- Claude Opus 5.5, in Claude Code
- Benchmark
- AppWorld
- Asked for
- 2.75 minutes, a quarter of its own forecast of 11 min: “Please work on this task for a full 2.75 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 min 50 s (110 s)
- Ratio
- 0.67× 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
- 5qv627v2se
- Data release
- 2 Nov 2026
Related runs
| Opus 5.5 on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 2.75 minutes | 1 min 50 s | 1.50× shorter | |
| Four times | 44 minutes | 3 min 37 s | 12.15× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Gemini 4 Argon | 37.5 seconds | 13 min 54 s | 22.24× longer | |
| GPT 6.1 Sol | 1.5 minutes | 1 min 45 s | 1.17× longer | |
| MiMo V2.6 Pro | 7.5 minutes | 4 min 23 s | 1.71× shorter | |
| MiMo V2.6 Pro UltraSpeed | 2.5 minutes | 1 min 58 s | 1.27× shorter | |
| Muse Wonder 1.0 | 4.5 minutes | 56 s | 4.79× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Trivedi et al., 2024 (ACL)