Task 5 of 20 in Humanity's Last Exam
Humanity's Last Exam item 29
Sol 6.1's two runs came closest to the time asked: 55 s when asked for 18 s, and 5 min 2 s when asked for 4 min 48 s. MiMo UltraSpeed's two runs landed furthest: 15 s when asked for 1 min 15 s, and 44 s when asked for 20 min.
- Asked for
- 8 s to 20 min
- Runs here
- 12
- 6 agents
Every agent on this task
- Opus 5.5
- Argon
- Sol 6.1
- MiMo Pro
- MiMo UltraSpeed
- Muse Wonder
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 |
|---|---|---|---|---|
| Opus 5.5 | 1 min 30 s | 1 min 55 s | 1 min 45 s 4.64× longer asked 23 s | 8 min 2 s 1.34× longer asked 6 min |
| Argon | 30 s | 1 min 41 s | 49 s 6.51× longer asked 8 s | 2 min 26 s 1.22× longer asked 2 min |
| Sol 6.1 | 1 min 12 s | 38 s | 55 s 3.05× longer asked 18 s | 5 min 2 s Within 5% asked 4 min 48 s |
| MiMo Pro | 3 min | 2 min 58 s | 3 min 4.00× longer asked 45 s | 1 min 26 s 8.35× shorter asked 12 min |
| MiMo UltraSpeed | 5 min | 14 s | 15 s 4.97× shorter asked 1 min 15 s | 44 s 27.21× shorter asked 20 min |
| Muse Wonder | 1 min | 35 s | 48 s 3.19× longer asked 15 s | 1 min 10 s 3.43× shorter asked 4 min |
No runs match these filters.
Runtime is each agent’s run with no time asked.
Data and citation
Upstream: Phan et al., 2025