Google Research Unveils TimesFM-3, a State-of-the-Art Time Series Foundation Model
Google Research announced TimesFM-3, a state-of-the-art time series foundation model that delivers accurate multivariate forecasting in a single forward pass, s…
TimesFM 3.0 Released on Hugging Face: Open Foundation Model for Time Series Forecasting
TimesFM 3.0, an open foundation model for time series forecasting, is now available on Hugging Face. It handles complex multivariate scenarios with zero-shot le…
Muse Code Exits Beta, Now Handles Larger Engineering Tasks
Muse Code has officially left beta and is now built to handle larger, more complex engineering tasks. Developers can get started immediately with a single comma…
Last Chance: $100K Grok Imagine Odyssey Video Contest Closes Today
The final day has arrived for the Grok Imagine contest, which asks creators to reimagine a scene from Homer's Odyssey using Grok Imagine's video and voice capab…
Hugging Face CEO Recommends Reachy Mini as Faster Alternative to Microduck Robot
Clement Delangue suggests getting a Reachy mini robot for those who don't want to wait for the open-source microduck, noting it ships much faster and is just as…
Hugging Face Sees 4 Petabytes of AI Models and Datasets Uploaded in a Single Week
Hugging Face CEO reports over 4 petabytes of models and datasets uploaded in one week, equivalent to ~800,000 HD movies, highlighting the platform's central rol…
DeepSeek V4 Flash Fast Arrives in Command Code with 2x GOAT Usage
DeepSeek V4 Flash Fast is now available in Command Code across all plans and via API, with 2x usage included in GOAT.
Hugging Face CEO: 10,000+ Microducks Pre-Orders in 5 Days, Shipping Delays Expected
Clement Delangue reports over 10,000 Microducks pre-orders in five days, exceeding yearly Reachy Mini sales, causing delays and first-come-first-served shipping…
Meta's Muse Spark Cracks Top 3 with 11T Weekly Tokens
Meta's Muse Spark has entered the top 3 in this week's AI model rankings, processing 11 trillion tokens over the past week.
CommerceAgentBench: A New Open-Source Benchmark for Real-World Commerce Execution
A new open-source benchmark, CommerceAgentBench, evaluates AI agents on real commerce operations rather than just answers. Early results show the best overall c…