Close Menu
  • AI
  • Content Creation
  • Tech
  • Robotics
AI-trends.todayAI-trends.today
  • AI
  • Content Creation
  • Tech
  • Robotics
Trending
  • A Group of Researchers from Columbia and CertiK Introduce TensorLean for Formal Verification of GPU Kernels
  • Collectively AI Discloses Immediate Leak After Coding Agent Uncovered Credential – Unite.AI
  • Dragos Finds Immediate Leaks and Stolen Keys in AI Provide Chains
  • Lean and Mathlib Communities Certify Claude-Assisted Riemann Proof – Unite.AI
  • AWS Routes HyperPod Inference Throughout Areas From One Endpoint – Unite.AI
  • DailyAI Briefing: A Lean-checked Riemann proof, plus 4 extra
  • Anthropic Shares Drop 8.5% in Second Session Throughout AI Selloff – Unite.AI
  • OpenAI Ships GPT-Stay-1.1 for Voice Brokers, Cuts Idle Value to $0.02/Min
AI-trends.todayAI-trends.today
Home»Tech»Google AI Releases TimesFM-3, A Multivariate Forecasting Model With 330M Parameters.

Google AI Releases TimesFM-3, A Multivariate Forecasting Model With 330M Parameters.

Tech By Gavin Wallace31/08/20264 Mins Read
Facebook Twitter LinkedIn Email
NVIDIA Releases Llama Nemotron Nano 4B: An Efficient Open Reasoning
NVIDIA Releases Llama Nemotron Nano 4B: An Efficient Open Reasoning
Share
Facebook Twitter LinkedIn Email

Google Research releases new Google Search. TimesFM-3The model is a time series foundation with over 330 millions parameters that can forecast multiple series at once. Each TimesFM checkpoint up to 2.5 was univariate, a single series with its own historical data, and nothing more. TimesFM-3 has been pretrained for multivariate prediction on over 1 trillion timepoints. It accepts multiple target, past covariates and past-future comorvariates without task-specific tweaking. The model is ranked first among all pretrained models in GIFT Evaluation, fev Bench, and TIME Leaderboard on point and probability metrics.

Is it deployable?

PartiallyThe TimesFM repository code is Apache-2.0But the TimesFM 3.0 weights ship under timesfm-non-commercial-license-v1.0. These are only for non-commercial and non-production purposes. It is available for benchmarking today. You can’t ship behind a forecast API.

What’s changed

Each TimesFM release up to 2.5 had a single-variable. This model predicted one series based on its history. The majority of real forecasting issues aren’t shaped this way. Google’s example Ice cream sales are affected by foot traffic, the weather, promotional campaigns, holidays, etc.

TimesFM-3 was pretrained to multivariate forecasting. It has 330,000,000 parameters and is pretrained with more than 1 trillion times points from real and synthetic series. There are three input types that work with zero-shot and no fine-tuning for specific tasks:

  • Multi-targets Forecasting jointly with quantile and point outputs.
  • Past covariatesOnly known from the past, for example, foot traffic.
  • Covariates of the past and futureThe future value of the object is known. For example, in a calendar for promotions.

Two types of architectural attention: first, patches.

Backbone is a transformer that only decodes. The points that are adjacent to each other are then grouped together into 32-step patches, and normalized by series. This prevents the use of scales with wildly different values. All tokens, including the target and past-covariates, are generated from one patch. Tokens for past and future covariates are generated using the lookahead technique: The current patch is combined with patches in the future, allowing the model to see events scheduled before they happen.

Tokens enter into a 2D Grid and then pass through two alternate attention mechanisms.

  1. Attention temporale causale The horizontal direction of the run is a straight line. The leakage is, therefore, strictly causal.
  2. Attention Full The vertical direction is used. Tokens are used to read every series of a certain timestep, and learn correlations between the different series.

The one forward pass is better than many

In earlier TimesFM versions, patches were decoded individually. This adds to latency and compute costs, as well as compounding errors. TimesFM-3 utilizes Contiguous Patch MaskingTiRex introduced the strategy of masking training times. The entire horizon is masked with placeholder tokens. There, targets and previous covariates will be masked. Future signals are still able to reach the model, as past covariates remain visible. Every masked patch of horizon is simultaneously filled with the alternating layers. Every horizon step, 9 quantiles are applied to each target, representing the 10th through 90th per centile.

Benchmarks

Google evaluates GIFT-Eval, fev-benchThe. TIME leaderboardTimesFM-2.5, Chronos-2 and the Toto 2.0 Family. TimesFM-3, among the pretrained models of foundations, is ranked highest on average for all three metrics, including point values and probabilistic measures. The package release notes Ranking #1 in TIME over 50 datasets with 98 evaluation Tasks.

Interactive explainer

‘;
}
h+=’

AI ar Google met
Share. Facebook Twitter LinkedIn Email
Avatar
Gavin Wallace

Related Posts

A Group of Researchers from Columbia and CertiK Introduce TensorLean for Formal Verification of GPU Kernels

22/11/2026

OpenAI Ships GPT-Stay-1.1 for Voice Brokers, Cuts Idle Value to $0.02/Min

21/11/2026

Meet Pondering Machines Inkling 2 Mini: 26B Looped MoE Runs on a 32GB Mac

21/11/2026

Baseten Releases Bison-1 and Groundhog-1, Topping Choice Mannequin Index

21/11/2026
Top News

Amazon Workers Issue Warning About Company’s ‘All-Costs-Justified’ Approach to AI Development

In China’s Hardware Capital, operating a humanoid with your body is a hot job.

I Thought I Knew Silicon Valley. I Was Mistaken

Digital Marketing Courses to Sell Digital Marketing Courses • AI Blog

Apple’s AI Ambitions Leaves Big Questions About Its Climate Goals

Load More
AI-Trends.Today

Your daily source of AI news and trends. Stay up to date with everything AI and automation!

X (Twitter) Instagram
Top Insights

Gemini Shouldn’t Want Human Evaluation to Learn Your Gmail

10/11/2026

A Coding Information to Google Analysis’s Kauldron: Configs That Are Plain Knowledge, Parts Wired by String, and a JAX Coach You Can Learn Finish to Finish

02/10/2026
Latest News

A Group of Researchers from Columbia and CertiK Introduce TensorLean for Formal Verification of GPU Kernels

22/11/2026

Collectively AI Discloses Immediate Leak After Coding Agent Uncovered Credential – Unite.AI

22/11/2026
X (Twitter) Instagram
  • Privacy Policy
  • Contact Us
  • Terms and Conditions
© 2026 AI-Trends.Today

Type above and press Enter to search. Press Esc to cancel.