New
Everything new in AI as it appears: launches announced by the labs, brand-new models, datasets, papers and open-source tools. Checked every minute or two. Launches feed · Trending · News
Live now
- Connecting…
Saturday 24 October
- Datasetgavinlaw/setground-docci-tasks
SetGround DOCCI artifacts Derived from Google DOCCI (Onoe et al., ECCV 2024), under CC BY 4.0. Fixed tasks and 560 pilot images. Fixed-model acquisition recovery snapshot
- DatasetChand0320/single-task-transfer-artifacts
Single-task Transfer SAILOR and Retrieval Artifacts Artifacts from the public code repository RahulSChand/single-task-transfer. Contents artifacts/encoders/: broad and pi
- DatasetGom-sy/transition-gen-world
transition-gen-world (world-model format) RoboCasa coffee3 ground truth (165 episodes) plus 144 world-model-generated transition clips, in the Cosmos Predict 2.5 multivie
- DatasetGom-sy/transition-gen-world-policy
transition-gen-world (GR00T policy format) RoboCasa coffee3 GT (165 episodes) plus 144 world-model-generated transition clips, labeled by the Inha kitchen3 IDM, in GR00T/
- Dataset
- DatasetSkinnyVIP/glp1-telehealth-real-price-index
GLP-1 Telehealth Real-Price Index — United States What patients actually pay each month across 11 GLP-1 telehealth programs after teaser rates, membership fees, dose surc
- Datasetkatf0/Farfetch.Product
Farfetch web scraped data About the website Farfetch operates in the dynamic and rapidly evolving E-commerce industry in the EMEA, particularly in the United Kingdom. Thi
- Dataset
- Dataset
- Dataset
- Datasetllitvak/Butterfinger_1_20261023_163001
This dataset was created using LeRobot. Dataset Structure meta/info.json: { "codebase_version": "v3.0", "fps": 30, "features": { "action": { "dtype": "float32", "shape":
- Dataset
- Datasettandiu/ai-infra-data-sample
TANDIU AI Infrastructure Data (Sample) TANDIU (钽铥) publishes daily-updated data on AI infrastructure: LLM API list prices, public benchmark scores, an LLM price index, GP
- Datasetefederici/smol-software-harbor
🛠️ smol-software-harbor 1,002 small, verifiable software-engineering tasks on 576 hand-written mini-codebases in 12 languages, for small coding agents: read, find the ri
- DatasetIhorShevchuk/IndicTTS_Punjabi_verified
IndicTTS Punjabi - verified pairing subset A corrected, training-ready subset of the SPRINGLab IndicTTS Punjabi corpus (AI4Bharat / IIT Madras), containing only the clips
- DatasetSkyeAv/biolink-amalgam
BiolinkAmalgam Biomedical entity synonym cliques built from BABEL compendia and synonyms (snapshot: local:/home/sgoetz/babel-snapshots/2026jul22), merged and arbitrated s
- Dataset
- DatasetSoroushVahidi/llm-serving-common-contract-m3
Common-contract LLM-serving scheduler results — M3 Release m3-v1.0.0. Author: Soroush Vahidi. Promoted from independently verified rc3 with unchanged result-table bytes.
- Datasetbenchflow/frontierphysics-icecube-pointsource-unbinned
IceCube data for neutrino point-source searches, years 2008-2018 (mirror) This is a byte-identical mirror of version 2.0 of the IceCube Collaboration's public point-sourc
- DatasetLyricBobby/Annoy-PyEdu-Rs-Raw
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- DatasetLyricBobby/Annoy-PyEdu-Rs
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- Datasetfusinjay/todd
Status: Complete for the scope described below. TODD original archives Original author: Haoping Xu, University of Toronto. Source: https://doi.org/10.5683/SP3/ZJJAJ3 (TOD
- Datasetfusinjay/transcg
Status: Complete for the scope described below. TransCG archive mirror Original authors: Hongjie Fang, Hao-Shu Fang, Sheng Xu, and Cewu Lu. Source: https://graspnet.net/t
- Datasetfusinjay/transpose
Status: Complete for the scope described below. TRansPose: All Objects and Test sequences Copyright SNU RPM Labs. Original authors: Jeongyun Kim, Myung-Hwan Jeon, Sangwoo
- Dataset
- DatasetYongYong/cis6280-stack-datasets-v2
CIS6280 stacking datasets, v2 (4-cube trajectories) 20,000 episodes of a dexterous-hand cube stack in the stacking-trajectories-v0.3 format, collected so that episodes co
- Datasethanfeng2006/excavator-motion
Introduction Excavator-motion is a large-scale dataset of excavator motion trajectories, encompassing three distinct excavator models. Each H5 file in the dataset records
- Datasetjackyk02/wildchat-qwen3.6-35b-a3b-fp8-embeddings-256k
WildChat: real user conversations with ChatGPT as chat trajectories: Qwen3.6-35B-A3B-FP8 state/action embeddings (state, action) embedding pairs from Qwen/Qwen3.6-35B-A3B
- Datasettakeforest/teamI_20261024_120057
This dataset was created using LeRobot. Dataset Structure meta/info.json: { "codebase_version": "v3.0", "fps": 30, "features": { "action": { "dtype": "float32", "names":
- Dataset
- Datasetformospeech/hakkatv_hanzawa
Dataset Information Corpus: hakkatv_hanzawa Status: private_and_closed Subset Stats Subset Train (Samples / Hours) Eval (Samples / Hours) hak-x-sixian 11,011 / 5.30 0 / 0
- Dataset
- Dataset
- Dataset
- Dataset
- Dataset
- Datasetlaion/Humaneness-Voice-Large-Alpha-Continuation-Listening
Humaneness Voice Large: additional 20/40/40 RL listening outputs 70 exact matched scripted takes from the separately identified continuation experiment. The released 1.4
- DatasetNoeFlandre/GeoReSeT-Corpus
GeoReSeT Corpus 96,675,303 sentences in 95 languages, tied to 3,704,364 OpenStreetMap polygons in 386 regions, each labelled for land-use / land-cover relevance by an LLM
- Dataset
- Dataset
Headlines and links come from the labs' own blogs, Hacker News and the GDELT news index, and from the sources our index already follows. We show a short excerpt and always link to the original. A launch reported by a news site rather than the lab is marked “Reported” until the lab itself announces it.