Constarium
← Search

Text · dataset · 2026

CivBench

Listed in Hugging Face Datasets

CivBench A benchmark of LLM agents playing full games of Civilization VI through an MCP (Model Context Protocol) server.

Description

Each run captures the full per-turn game state, every tool call the agent issued, and the agent's own structured reflections. Configs tables — curated parquet tables, one row per logical record.

Use this for analysis: datasets.load_dataset("civbench/civbench-v1", "tables", split="games"). raw — byte-identical mirror of the on-disk telemetry JSONL… See the full description on the dataset page: huggingface.co/datasets/civbench/civbench-v1.

Links

Where it is published

Catalogue records · 1

Topics

Stated by source
other · tabular · text
Provenance · 1 source records, 11 field assertions
SourceKeyLast seenRaw
Hugging Face Datasetscivbench/civbench-v114 d agoJSON v1
FieldAssertionExtractorEvidence
access_levelsource · Hugging Faceconnector:huggingface@1.0.0/gated
concepts[field].local:field:computer-science-aimapping · Hugging Faceconnector:huggingface@1.0.0
concepts[modality].hf_modality:tabularsource · Hugging Faceconnector:huggingface@1.0.0
concepts[modality].hf_modality:textsource · Hugging Faceconnector:huggingface@1.0.0
concepts[task].hf_task:othersource · Hugging Faceconnector:huggingface@1.0.0/tags[task_categories:*]
created_datesource · Hugging Faceconnector:huggingface@1.0.0
descriptionsource · Hugging Faceconnector:huggingface@1.0.0/description
licensesource · Hugging Faceconnector:huggingface@1.0.0/tags[license:*]
publication_datesource · Hugging Faceconnector:huggingface@1.0.0
titlesource · Hugging Faceconnector:huggingface@1.0.0/id
updated_datesource · Hugging Faceconnector:huggingface@1.0.0