FlyPig23/memory_dci
Memory DCI Two benchmark datasets, organized as parallel folders. Code and runnable instructions live in FlyPig23/Memory_DCI. Benchmark Frozen split Contents Directory WildClawBench 36 training / 24 test 432 training trajectories, 381 images, task inputs, 72 V1/V1.1/V5 historical results wildclaw_bench/ TerminalBench 2.1 53 training / 36 test 89 official task packages, 5,785 training trajectory bodies, exact task order and provenance terminal_bench_2_1/ Each… See the full description on the dataset page: https://huggingface.co/datasets/FlyPig23/memory_dci.
Memory DCI
Two benchmark datasets, organized as parallel folders. Code and runnable instructions live in FlyPig23/Memory_DCI.
Each folder contains its own README, manifests, task split, sources and redaction records. These are distinct partitions; no tasks are resampled by downloading them. Both downloaders pin a publication and verify archive/per-file SHA256.
python3 -m experiment.benchmarks.wildclaw_bench.scripts.download_data --include-evidence
python3 -m experiment.benchmarks.terminal_bench_2_1.scripts.download_dataThe 2026-10-05 layout migration moved the original root WildClaw files into wildclaw_bench/ and changed their restored prefix to experiment/benchmarks/wildclaw_bench/. Task contents, task split/order, trajectory bodies, existing results and scores did not change. Three operational WildClaw manifests have updated file paths; original hashes and mappings are recorded. The TerminalBench archives are unchanged. The previous root layout remains available in commit b31c851fc4c09108ace8fbdf80dc5c1b1b0c396d.
No credentials, active experiment directories, private runtime environments or retired experiment data are published. Earlier redactions remain documented. Source-specific licenses and third-party rights are described in each folder; no blanket license is asserted over all embedded assets or Hub trajectories.
