Terminal-Bench

Tools & Infrastructure

Terminal agent performance. GPT-5.4 leads at 77.3%.

🌐 Visit website ↗

🔗 Similar tools

OpenAI Codex CLI ★ 121.7k

OpenAI Codex CLI is an open-source terminal coding agent written in Rust (Apache-2.0, 82K+…

Coding Agents
Harbor ★ 5.0k

Harbor is an Apache-2.0 framework, developed by Stanford and the Laude Institute, for eval…
MAI-Code-1.1-Flash

MAI-Code-1.1-Flash is a production coding assistant that delivers higher-quality code than…
SWE-bench Pro

SWE-bench Pro is a repository-level software engineering benchmark that extends the Verifi…
SWE-bench ★ 5.8k

SWE-bench is a benchmark that evaluates LLM-based systems on their ability to resolve real…
τ²-bench (tau2-bench) v1.0.1 ★ 2.0k

τ²-bench v1.0.1 is Sierra Research's benchmark for evaluating tool-agent-user interaction …