datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
berkeley_rpt_lerobotThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.0",
"robot_type": "franka",
"total_episodes": 908,
"total_frames": 392578,
"total_tasks": 4,
"total_videos": 908,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:908"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/IPEC-COMMUNITY/berkeley_rpt_lerobot.exp_rpt_multifile-a0-base-pass8-300
TaskTrove A0 base pass@8 cohort
This repository contains the 300 extracted Harbor tasks selected for the A0
base-model pass@8 evaluation. The source is DCAgent/exp_rpt_multifile at
revision 70527e80ee7497c800ea0f9bad90b87423c784c9.
cohort.json records the deterministic random selection algorithm, seed, and
complete task list. The task directories are the unmodified archives from the
source dataset.
berkeley_rpt_rawberkeley_rptThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.0",
"robot_type": "unknown",
"total_episodes": 908,
"total_frames": 392578,
"total_tasks": 4,
"total_videos": 908,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:908"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/lerobot/berkeley_rpt.exp_rpt_pymethods2test-large-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_pymethods2test-large-qwen3.5-122b-131k-opencode-traces.a3-rl-DCAgent_exp_rpt_pymethods2test-largerl_rl-conf_24GP_base-yaml_mode-path_r2eg-nl2b-stac-bugs-fixt_trai-data_exp_rpt_stac-php-largexp_rpt_pr-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_pr-qwen3.5-122b-131k-opencode-traces.rl__24GPU_base__exp_rpt_issue__Qwen3-8Ba3-rl-laion_exp_rpt_codenet-python-v2rl__24GPU_base__exp_rpt_scaffold__Qwen3-8Brl_rl-conf_24GP_base-yaml_mode-path_r2eg-nl2b-stac-bugs-fixt_trai-data_exp_rpt_stac-self-largrl__24GPU_base__exp_rpt_multifile__Qwen3-8Brl_rl-conf_24GP_base-yaml_mode-path_r2eg-nl2b-stac-bugs-fixt_trai-data_exp_rpt_pyme-larga3-rl-DCAgent_exp_rpt_curriculum-mediumexp_rpt_methods2test-large-v3-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_methods2test-large-v3-qwen3.5-122b-131k-opencode-traces.rl_rl-conf_24GP_base-yaml_mode-path_r2eg-nl2b-stac-bugs-fixt-agai_trai-data_exp_rpt_stac-rustexp_rpt_ghactions-v3-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_ghactions-v3-qwen3.5-122b-131k-opencode-traces.exp_rpt_unitsyn-python-large-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_unitsyn-python-large-qwen3.5-122b-131k-opencode-traces.rl_rl-conf_24GP_base_noth-yaml_mode-path_r2eg-nl2b-stac-bugs_trai-data_exp_rpt_stac-self-largrl__40GPU_base_32b__exp_rpt_nemotron-bash__Qwen3-32Bexp_rpt_crosscodeeval-java-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_crosscodeeval-java-qwen3.5-122b-131k-opencode-traces.RPToolkit-demo-datasetRPToolkit is a data generation pipeline, part of Augmentoolkit, that generates synthetic RP sessions inspired by input stories. Basically: feed in Lord of the Rings, get out high fantasy adventure RPs.
This dataset, containing over a million trainable tokens across around 1000 RP sessions, is meant to showcase the capabilities of this pipeline.
The input texts used were: a variety of myths and classic stories from Gutenberg; the first few chapters of some miscellaneous webnovels and… See the full description on the dataset page: https://huggingface.co/datasets/Heralax/RPToolkit-demo-dataset.DeKH
DeKH — German Hospital Dataset
DeKH (German Hospital Dataset) is a multi-scene dataset of real hospital environments comprising high-resolution 3D point clouds with semantic annotations and ground-truth IFC BIM models. It is introduced alongside BIMStruct3D, a fully automated hybrid Scan-to-BIM pipeline accepted at EC3 2026.
Dataset Contents
The dataset covers four distinct hospital scenes across three buildings:
Buildings/
├── A/
│ ├── 1st_floor/
│ │… See the full description on the dataset page: https://huggingface.co/datasets/RPTU-FGMB/DeKH.exp_rpt_e2egit-large-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_e2egit-large-qwen3.5-122b-131k-opencode-traces.rl__24GPU_base__exp_rpt_pymethods2test-large__qwen3base-GLM-4_7-swsec-edgar-rptexp_rpt_crosscodeeval-csharp-v4-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_crosscodeeval-csharp-v4-qwen3.5-122b-131k-opencode-traces.rl_rl-conf_24GP_base_noth-yaml_mode-path_r2eg-nl2b-stac-bugs_trai-data_exp_rpt_pyme-largexp_rpt_stack-pytest-large-qwen3.5-122b-131k-opencode-traces
Agent trace dataset
Decoding the literal token IDs
The prompt_token_ids / completion_token_ids / logprobs columns are the
verbatim tokens the serving engine emitted, stored PER AGENT STEP as a
list-of-lists (one inner list per turn). To turn them back into text you MUST
use the exact tokenizer the model was served with — a generic same-family
tokenizer will decode word tokens to garbage.
Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8
from transformers… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/exp_rpt_stack-pytest-large-qwen3.5-122b-131k-opencode-traces.