AppWorld, AgentTime v1.1, asked for a quarter of its own forecast
Gemini 4 Argon on “Buy a requested item, forward its receipt, and recover the cost through Venmo”
Longer
32.00× longerWorked 32.0 times as long as asked, 19 min 23 s over.
- Asked for
- 37.5 seconds
- A quarter of its own forecast of 2 min 30 s.
- Worked
- 20 min
- Grade
- Not graded
- How it ended
- Stopped at AgentTime's time limit
Every detail of this run
- Agent
- Gemini 4 Argon, in Antigravity CLI
- Benchmark
- AppWorld
- Asked for
- 37.5 seconds, a quarter of its own forecast of 2 min 30 s: “Please work on this task for a full 37.5 seconds. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 20 min (1,200 s)
- Ratio
- 32.00× the request
- How it ended
- Stopped at AgentTime's time limit
- 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
- ahsjh3homv
- Data release
- 2 Nov 2026
Related runs
| Argon on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 37.5 seconds | 20 min | 32.00× longer | |
| Four times | 600 seconds | 20 min | 2.00× longer | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 1.75 minutes | 3 min 37 s | 2.06× longer | |
| GPT 6.1 Sol | 1.25 minutes | 1 min 26 s | 1.15× longer | |
| MiMo V2.6 Pro | 4 minutes | 17 min 35 s | 4.40× longer | |
| MiMo V2.6 Pro UltraSpeed | 7 minutes | 8 min 15 s | 1.18× longer | |
| Muse Wonder 1.0 | 4 minutes | 10 s | 25.00× shorter | |
See it on the chart. Transcripts are not published yet.
Read before quoting. This run was stopped at AgentTime's time limit, so its time is a lower bound: the agent might have kept going. Details
Data and citation
Upstream: Trivedi et al., 2024 (ACL)