datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
forge-3b-pretrain-data
FORGE-3B Pretraining Data
Tokenized and packed pretraining data for the FORGE-3B language model.
Stats
Total tokens: 51.4070B
Domains: 10/10
Sequence length: 2048 tokens
Format: .npy shards of shape (N, 2048) with dtype uint32
Tokenizer: CRAYON (xerv-crayon, standard profile)
Domain Breakdown
Domain
Weight
Tokens (B)
Status
fineweb_edu
30%
15.0008
✓
thestack
16%
8.0011
✓
wikipedia
8%
4.2791
✓
openwebmath
8%
3.9654
✓
books
7%… See the full description on the dataset page: https://huggingface.co/datasets/Phase-Technologies/forge-3b-pretrain-data.wmdp-bio-forget-corpusforgespectrum-114k
ForgeSpectrum (v3) — AI-Generated Image Detection with Reasoning Traces
ForgeSpectrum is a multi-domain corpus for AI-generated / manipulated image
detection, annotated by Gemini-2.5-Pro with structured forensic reasoning traces
(<fast>/<planning>/<reasoning>/<reflection>/<conclusion> patterns) plus per-image
attributes and suspicious-region notes.
v3 — what changed
v3 is the cleaned, balanced release:
3 domains: faces, scenes, id_cards (docs and scene_text… See the full description on the dataset page: https://huggingface.co/datasets/hardiksharma6555/forgespectrum-114k.sd-webui-forge-classicStable Diffusion WebUI Forge - Classic
[ Classic | Neo ]
Stable Diffusion WebUI Forge is a platform on top of the original Stable Diffusion WebUI by AUTOMATIC1111, to make development easier, optimize resource management, speed up inference, and study experimental features.
The name "Forge" is inspired by "Minecraft Forge". This project aims to become the Forge of Stable Diffusion WebUI.
- lllyasviel
(paraphrased)
"Classic" mainly serves as an archive for the "previous" version of… See the full description on the dataset page: https://huggingface.co/datasets/WhiteAiZ/sd-webui-forge-classic.Heap-Forge
DISCLAIMER
This represents only a subset of the final dataset (taking into account the new HuggingFace LFS storage limits). The complete dataset will be released following the camera-ready submission of our paper.
The Heap Dataset
We develop The Heap, a new contamination-free multilingual code dataset comprising 57 languages, which facilitates LLM evaluation reproducibility. The reproduction packge can be found here.
Is your code in The Heap?
An opt-out… See the full description on the dataset page: https://huggingface.co/datasets/WizzF/Heap-Forge.Forge
Forge — HEGEMON (territorial.io vision AI)
Start here: HANDOFF.md — self-contained, rewritten 2026-09-17. It covers the
mission, credentials, the current best model configuration, every verified number, every falsified
lever, and the prioritized next steps. A capable agent can resume with this file plus this repo and
nothing else.
Live status log: REPORT.md — newest entry at the top, plain language, honest
about failures. Read it before believing any older document, including… See the full description on the dataset page: https://huggingface.co/datasets/amer224/Forge.crossfps-lmdb-2607egoexo-forge-egodexswe-forge
SWE-Forge Dataset
20 validated tasks for evaluating software engineering agents.
Each task contains:
workspace.yaml - Task configuration (repo, commits, install commands, test commands)
patch.diff - The ground-truth patch
tests/ - Generated test files (fail before patch, pass after)
evaluate.sh - Binary evaluator (score 0 or 1)
Docker Images
Pre-built images on Docker Hub: platformnetwork/swe-forge:<task_id>
Each image has the repo cloned at base_commit with… See the full description on the dataset page: https://huggingface.co/datasets/CortexLM/swe-forge.forge-arm-pixels
physicalai-bmi/forge-arm-pixels
Real MuJoCo pixels captured live from the Institute's in-browser Forge arm (WebGPU),
paired with the action the released state-checkpoint took. This is the exact training
set behind physicalai-bmi/nano-vla-pixels.
2,500 frames across 128 reaches, frames/f#####.png (the rendered MuJoCo arm, 844×520).
meta.json — per-frame { i, act:[3], obs:[7], reaches }; act is the 3-D joint-delta action, reaches is the episode index (use it for an episode-level… See the full description on the dataset page: https://huggingface.co/datasets/physicalai-bmi/forge-arm-pixels.ForgeWebSearch-BenchForgeWM-data
ForgeWM Training Data
Pre-encoded latents for training ForgeWM (arXiv:2608.14022).
Project Page: https://asdfo123.github.io/ForgeWM/
This is a re-packaging of the GameFactory
GF-Minecraft dataset, encoded into Wan2.1 VAE latents and sharded as LMDB for
direct use by ForgeWM training scripts. The underlying gameplay videos are
from GameFactory — we just made them training-ready.
Quick Start
huggingface-cli download ForgeWM/ForgeWM-data \
--local-dir… See the full description on the dataset page: https://huggingface.co/datasets/ForgeWM/ForgeWM-data.Hephaestus-ForgeOpen-Omega-Forge-1M
Open-Omega-Forge-1M
Open-Omega-Forge-1M is a carefully curated and optimized collection derived from multiple high-quality datasets, specifically designed to enhance reasoning capabilities across mathematical, scientific, and coding domains. This dataset represents a focused subset that maintains the quality and diversity of reasoning patterns while providing a more manageable size for training and evaluation. A high-quality, compact reasoning dataset designed for mathematics… See the full description on the dataset page: https://huggingface.co/datasets/prithivMLmods/Open-Omega-Forge-1M.FORGE-Curated
[!IMPORTANT]
IMPORTANT NOTE
Due to Huggingface storage limitations, the dataset files are not complete, and will not be maintained here, please refer to our Github Repo: https://github.com/shenyimings/FORGE-Curated
FORGE Curated: A Curated EVM Smart Contracts Vulnerability Dataset
FORGE Curated is a high-quality subset of the FORGE dataset, specifically designed to support advanced research in smart contract security, including AI-based auditing, vulnerability analysis, etc.… See the full description on the dataset page: https://huggingface.co/datasets/Seuilping/FORGE-Curated.BLUR
BLUR: A Benchmark for LLM Unlearning Robust to Forget-Retain Overlap
The BLUR dataset expands on existing unlearning benchmarks by providing harder evaluation tasks, combined forget/retain queries, and relearning datasets of varying degrees of difficulty. Despite the benign nature of the queries considered, we find that the performance of existing methods drops significantly when evaluated on BLUR, with simple approaches performing better on average than more recent methods.… See the full description on the dataset page: https://huggingface.co/datasets/forgelab/BLUR.forge-failure-v1
Forge failure v1 — the episodes the expert lost
The failed episodes from the same collection that produced
angledusgar/forge-v1: identical policies,
identical environments, identical recording — only the outcome differs. Same schema, so the two
sets can be mixed or contrasted directly.
They are published apart from the successes on purpose. Imitation learning on failed trajectories
teaches failing, so a behaviour-cloning consumer should never receive them by accident; a failure… See the full description on the dataset page: https://huggingface.co/datasets/angledusgar/forge-failure-v1.wmdp-cyber-forget-corpuslemonilia-Roleplay-Forums
Roleplay Forums 2023-04
As part of a probably misguided effort for gathering useful data for a roleplay finetune, around April 2023 I scraped several of the largest or most popular English-language roleplaying forums publicly known, for a total of about 47 GB of data (uncompressed, including HTML tags and metadata), and posted the files on a public Rentry page. I handpicked and cleaned a tiny fraction of that data for LimaRP, among other things. Other people posted partially cleaned… See the full description on the dataset page: https://huggingface.co/datasets/NewEden-Forge/lemonilia-Roleplay-Forums.forge-476285d0-a71c-45b1-9968-d0ac2595aeb5claude-oss-sftforge-7b240721-8f21-440a-8cf5-fd26fb4fe718forge-9ba77a5c-4589-4f51-99d7-baf52b3a848cforge-intentdata
Forge Intent Dataset
Version: 1.0.0
FORGE-Bench
FORGE Bench generated videos
This dataset release contains generated videos whose retained provenance has
been accepted for publication.
Currently published collections:
Seedance 2.5
Veo 3.1 Fast
A previously evaluated 500-video candidate collection is not part of this
release because its generator identity cannot be independently verified from
retained submission, query, or retrieval records. Neither that collection nor
its scores should be attributed to a named model.
See… See the full description on the dataset page: https://huggingface.co/datasets/aaaabcd/FORGE-Bench.forge-extensionsforge
FORGE: Fine-grained Multimodal Evaluation for Manufacturing Scenarios
🌐 Website | 📑 Paper | 💻 Code | 🤗 Dataset
Quick Start
from datasets import load_dataset
ds = load_dataset("AI4Manufacturing/forge", "task1_three_view", split="train")
print(ds[0].keys())
ds[0]["test_image"] # PIL Image
Configs
Core… See the full description on the dataset page: https://huggingface.co/datasets/AI4Manufacturing/forge.forge-v1
Forge v1 — contact-rich manipulation with force/torque, 3 tasks
Expert demonstrations for the three Isaac Lab FORGE
tasks, collected from trained rl_games PPO policies. Successful episodes only; the failures from
the same collection are published separately as
angledusgar/forge-failure-v1.
Beyond the usual LIBERO-shaped fields this dataset carries the wrist force/torque signal, which
is what makes contact-rich insertion legible: the peg/hole clearance is 0.114 mm, far below what… See the full description on the dataset page: https://huggingface.co/datasets/angledusgar/forge-v1.mole
Mole: Detecting Insider Threats in AI Agents
Corpora for Mole, a benchmark for detecting
insider threats by AI agents in a simulated frontier AI lab.
Paper: arXiv:2609.06966.
150 AI-operated accounts share nine stateful services (GitLab, ownCloud, email, chat, a
Plane-style issue tracker, a model registry, an eval server, a secrets store, and an org
directory) over 30 workdays. On some account-days an account pursues a concealed malicious
objective alongside its routine work.… See the full description on the dataset page: https://huggingface.co/datasets/forgelab/mole.ForgeWM-GT-1000
ForgeWM Minecraft evaluation data
This private repository contains two separate evaluation inputs. They are kept
in different top-level directories to avoid confusing the paired Full-1000
conditioning frames with the independent 77-scene constant-action suite.
Layout
gt_1000/
videos/ # 1,000 GT videos, 81 frames, 640x352, 12 FPS
first_frames/ # 1,000 conditioning frames paired with the videos
actions/ # aligned… See the full description on the dataset page: https://huggingface.co/datasets/ForgeWM/ForgeWM-GT-1000.