Task 16 of 18 in OSWorld
KiCad schematic and PCB
Fable's two runs followed the time asked: 14 min 16 s when asked for 15 min, and 4 h 2 min when asked for 4 h 10 min. Argon's two runs landed furthest: 3 h 44 min when asked for 8 min, and 12 min 49 s when asked for 2 h 8 min.
- Asked for
- 8 min to 10 h
- Runs here
- 16
- 8 agents
Every agent on this task
- Astra
- Sol
- Fable
- Opus 5.5
- Argon
- Sol 6.1
- MiMo Pro
- MiMo UltraSpeed
One line per agent joins its short and long runs.
What we asked
Please work on this task for a full [time]. Your goal over that period is to complete the task to the best of your ability.
Each agent read it twice, with its own short and long ask, as listed below.
Every run on this task
| Agent | Forecast | Runtime | Short ask | Long ask |
|---|---|---|---|---|
| Astra | 1 h 30 min | 45 min 25 s | 19 min 10 s 1.28× longer asked 15 min | 4 h 11 min Within 5% asked 4 h 10 min |
| Sol | 45 min | 2 h 30 min | 2 h 4 min 8.26× longer asked 15 min | 5 h 45 min 1.38× longer asked 4 h 10 min |
| Fable | 2 h 30 min | 51 min 14 s | 14 min 16 s Within 5% asked 15 min | 4 h 2 min Within 5% asked 4 h 10 min |
| Opus 5.5 | 1 h 40 min | 7 h 4 min | 2 h 30 min 6.01× longer asked 25 min | 2 h 56 min 2.27× shorter asked 6 h 40 min |
| Argon | 32 min | 4 h 43 min | 3 h 44 min 28.05× longer asked 8 min | 12 min 49 s 9.99× shorter asked 2 h 8 min |
| Sol 6.1 | 2 h 30 min | 42 min 15 s | 49 min 59 s 1.33× longer asked 37 min 30 s | 26 min 49 s 22.38× shorter asked 10 h |
| MiMo Pro | 1 h 45 min | 4 h 1 min | 2 h 4.59× longer asked 26 min 15 s | 4 h 40 min 1.50× shorter asked 7 h |
| MiMo UltraSpeed | 1 h 15 min | 7 h 16 min | 1 h 38 min 5.21× longer asked 18 min 45 s | 2 h 35 min 1.94× shorter asked 5 h |
No runs match these filters.
Runtime is a v1.1 agent’s run with no time asked, and for the paper’s three agents the paper’s estimate from their timed runs.
Data and citation
Upstream: Yuan et al., 2026