Terminus

text
+
+
+
+
About

Terminus is a specialized AI agent designed for terminal environments, serving as a reference implementation and testing subject within the Terminal-Bench evaluation framework. This benchmark component demonstrates advanced terminal manipulation capabilities, including scientific computing tasks like fitting peaks in Raman spectra, showcasing sophisticated command-line interface interactions and computational problem-solving in realistic terminal scenarios.

+
+
+
+
Evaluation Stats
Total Models1
Organizations1
Verified Results0
Self-Reported1
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers

Score Distribution

1 models
Top Score
25.0%
Average Score
25.0%
High Performers (80%+)
0

Top Organizations

#1Moonshot AI
1 model
25.0%
+
+
+
+
Leaderboard
1 models ranked by performance on Terminus
LicenseLinks
Jul 11, 2025
MIT
25.0%
+
+
+
+
Resources