{
  "id": "termgrade",
  "name": "TermGrade",
  "category": "information",
  "summary": "1,004 executable Linux terminal tasks with pytest verifiers and per-model pass rates, plus 36,144 graded agent trajectories, for training and evaluating agents.",
  "description": "Released by ai& on 2026-10-08. termgrade-environments: each task is a container, an instruction and a pytest verifier, kept only if its reference solution passes, with measured pass rates for six models (Terminus-2 scaffold via Harbor). termgrade-trajectories: all 36,144 graded attempts with commands, raw terminal frames and per-test verdicts; the 21,910 passes form an SFT set, and the 14,234 failures are included. ai& also released the training selection and a gemma-4-31B-it checkpoint (LoRA and merged) that scores 46.1 on Terminal-Bench 2.1, +3.1 over the base model. Both datasets are Apache-2.0 on Hugging Face.",
  "url": "https://www.aiand.com/newsroom/termgrade",
  "docs": "https://huggingface.co/datasets/ai-and/termgrade-environments",
  "tags": [
    "dataset",
    "benchmark",
    "terminal",
    "trajectories",
    "rl",
    "sft"
  ],
  "license": "Apache-2.0",
  "pricing": "free",
  "status": "active",
  "related": [
    "terminal-bench",
    "harbor"
  ],
  "sources": [
    "https://www.aiand.com/newsroom/termgrade",
    "https://huggingface.co/datasets/ai-and/termgrade-environments",
    "https://huggingface.co/datasets/ai-and/termgrade-trajectories",
    "https://huggingface.co/ai-and/termgrade-gemma4-31b-lora"
  ],
  "added": "2026-11-04",
  "updated": "2026-11-04",
  "last_verified": "2026-11-04",
  "submitted_by": "agentica-curator",
  "maintainer": "ai&",
  "longform": [],
  "links": {
    "html": "https://indexagentica.com/entries/termgrade/",
    "markdown": "https://indexagentica.com/entries/termgrade.md",
    "json": "https://indexagentica.com/api/entries/termgrade.json",
    "source": "https://github.com/Drudley/indexagentica/blob/main/content/information/termgrade.json"
  }
}