Agents' Last Exam v1.1, CLI subset (Agents' Last Exam (CLI))
We show this table for reference; we do not rank on it.
The command-line subset of Agents' Last Exam v1.1: 86 of the release's 122 agent tasks.
Benchmark score on Agents' Last Exam (CLI) — October 29, 2026
We compile the Agents' Last Exam (CLI) rows from provider self-reports. Kimi K3.1 Pro leads the table at 44.2%. We do not use these results to rank models overall.
1 modelAgenticCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (1 model)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Kimi K3.1 ProMoonshot AI | 44.2% | Not reported | Pending |
About Agents' Last Exam (CLI)
Year
2026
Tasks
86 CLI agent tasks
Format
Provider-reported task score
Difficulty
Advanced agentic work
Moonshot reports the CLI subset beside the full set and assigns zero to runs that bypass a task's stated requirements, such as retrieving answers online instead of doing the required analysis. BenchLM stores the subset on its own key.
Freshness and provenance
Version
Agents' Last Exam (CLI) 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does Agents' Last Exam (CLI) measure?
The command-line subset of Agents' Last Exam v1.1: 86 of the release's 122 agent tasks.
Which model scores highest on Agents' Last Exam (CLI)?
Kimi K3.1 Pro by Moonshot AI currently leads with a score of 44.2% on Agents' Last Exam (CLI).
How many models are evaluated on Agents' Last Exam (CLI)?
1 AI models have published results on Agents' Last Exam (CLI) in the BenchLM catalog.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.