FreeCAD 3D modeling benchmark runs
A table of results from 1,000 runs of AI agents on 100 parametric CAD generation tasks from text.
Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.
Answered by Gemini 4 · the chat is kept only in this tab
A table of results from 1,000 runs of AI agents on 100 parametric CAD generation tasks from text.
A set of 1 to 10 million images and English text annotations for training text-to-image models in text rendering.
10K to 100K English questions on long videos with a hard subset.
A corrected version of Stanford Alpaca for instruction fine-tuning in English.
Hundreds of thousands of audio clips from 14K hours of speech with timestamp annotations and summaries.
An index of direct links and bibliographic metadata for open scientific PDFs in English and German.
A decontaminated version of evol-codealpaca-v1 for training code generation without test leaks.
16 kHz Russian speech audio recordings with manual annotations from the crowdsourced part of the Golos corpus for ASR tasks.
Classic dataset with 54,306 leaf images across 14 crops for training plant disease recognition models.
Part of a massive 1 trillion token multimodal dataset for pre-training models in document and page understanding.
Large corpus of the top 10% FineWeb2 documents across 100 languages for high-quality language model pretraining.
An English dataset of 10K–100K examples featuring scene descriptions, trajectories, and VQA tasks for embodied AI.
A large English FinePhrase synthetic dataset for training LLMs in logic and instruction following.
A large heterogeneous English speech corpus with transcriptions and ready-made subsets for training ASR models.
Tabular correction tables for microrobot modeling using the boundary element method.
An English question-answering dataset (10K–100K examples) to evaluate language models' understanding of everyday physics.
A cleaned 18-language text corpus (1B to 10B records) with a focus on low-resource African languages.
Multimodal image-text dataset for testing vision-language models (VLMs).
Over 366 million audio clips with captions used to train LAION CLAP2 models.
A set of over 27,000 annotated images for object segmentation tasks.
Tabular dataset with cryptographically signed benchmark results of open models across various metrics.
A dataset of annotated CCTV clips for evaluating VLMs in industrial safety.
7,931 voice embeddings (X-vectors) for 7 speakers, extracted from the CMU ARCTIC speech dataset.
A robotics dataset containing video, tabular, and time-series data created via LeRobot.