0

Terminal-Bench (Hard)

Frontier

Hard subset of Terminal-Bench - long-horizon shell + filesystem tasks.

Published
May 2026
Canonical
tbench.ai

Cite

Notes

Only stored in your browser.

Attribution

Leaderboard scores
AA
Attribution policy →

Top score 62.9% by Claude Fable 5 (batch) - 311 models reporting (63 frontier)

Score history

311
0%25%50%75%100%Mar 24Sep 24Mar 25Sep 25Mar 26Claude 3 HaikuLlama 3.1 70B Instructo1 (batch)Grok 3 miniGrok 4Gemini 3 ProGPT-5.3-CodexClaude Opus 4.7 (batch)Claude Fable 5 (batch)

Top models

311
Terminal-Bench (Hard)Bar chart with 21 bars. Highest value: Claude Fable 5 (batch) at 62.9.
21 models

FAQ

What is Terminal-Bench (Hard)?
Hard subset of Terminal-Bench - long-horizon shell + filesystem tasks.
What is the current top score on Terminal-Bench (Hard)?
The top reported score is 62.9% by Claude Fable 5 (batch), across 311 models reporting (63 from frontier labs).