Skip to main content
ToolPotion
AI Advisor
Submit a Tool (Free)

TimeH2O-v1

TimeH2O-v1 is a 108M-parameter time series foundation model designed for zero-shot probabilistic forecasting. It supports covariates and delivers state-of-the-art results on benchmarks like GIFT-Eval and fev-bench. This model is ideal for applications requiring high accuracy in predicting future time-dependent data points, offering robust performance for various forecasting tasks.

View Model
Share
View Model
Share

TimeH2O-v1 is a sophisticated time series foundation model developed by H2O.ai, featuring 108 million parameters. It is specifically engineered for zero-shot probabilistic forecasting, meaning it can make predictions on unseen data without requiring prior fine-tuning on that specific dataset. A key capability of TimeH2O-v1 is its support for covariates, allowing it to incorporate additional relevant variables alongside the primary time series data to enhance prediction accuracy.

Read More

The model has demonstrated state-of-the-art performance across multiple benchmarks. On GIFT-Eval, it achieved a CRPS (Continuous Ranked Probability Score) of 0.441 and a MASE (Mean Absolute Scaled Error) of 0.654, outperforming models like Timestral-209M, t0-gamma, TimesFM-3, and Chronos-2. Similarly, on fev-bench, TimeH2O-v1 recorded an SQL skill score of 50.83 and a MASE skill score of 39.95, again surpassing its competitors.

TimeH2O-v1 is part of the broader TimeH2O family, which also includes TimeH2O-v1-tiny, a smaller 5.8M-parameter model optimized for maximum throughput rather than peak accuracy. The v1 model is best suited for scenarios where maximizing prediction accuracy is the primary goal. It can process up to 16,384 time steps of context and offers an unlimited prediction horizon, outputting 9 quantiles (from 0.1 to 0.9) for probabilistic forecasts. The model accepts target series with optional covariates as inputs.

Access to TimeH2O-v1 on Hugging Face requires users to agree to share their Hugging Face information and accept specific conditions, as it is licensed for non-commercial use. The model was trained on a diverse set of datasets, with careful exclusion of any data overlapping with fev-bench, GIFT-Eval, and TIME to ensure unbiased benchmark results. Training data sources include autogluon/chronos_datasets, Salesforce/GiftEvalPretrain, Salesforce/lotsa_data, hq-bench/quito-corpus, Dingdong-Inc/FreshRetailNet-LT, Azure Functions 2019 trace, Iowa Liquor Sales, and BLS Average Prices, each with their respective licenses.

  • CategoryAI Models & LLMs
  • DeploymentCloud
  • PlatformsHugging Face

  • 108M parameters for robust forecasting

  • Zero-shot probabilistic forecasting capabilities

  • Support for covariates to enhance prediction accuracy

  • State-of-the-art benchmark results on GIFT-Eval and fev-bench

  • Context window of up to 16,384 time steps

  • Unlimited prediction horizon

  • Outputs 9 quantiles (0.1 to 0.9) for probabilistic forecasts

  • Optimized for maximum accuracy in time series predictions

  • Non-commercial license for usage

  • Exclusion of benchmark-overlapping data from training

  • Part of the TimeH2O model family

  • Publicly accessible on Hugging Face with conditions

  1. Access Model: Navigate to the TimeH2O-v1 page on Hugging Face.

  2. Accept Conditions: Review and accept the terms and conditions to access the model's files and content.

  3. Log In/Sign Up: Log in to your Hugging Face account or sign up if you don't have one.

  4. Download Files: Download the model files and associated resources.

  5. Set Up Environment: Configure your development environment with necessary dependencies for using Hugging Face models.

  6. Integrate Model: Integrate TimeH2O-v1 into your application or workflow for time series forecasting.

  7. Provide Inputs: Supply target series data along with optional covariates to the model.

  8. Generate Forecasts: Utilize the model to generate zero-shot probabilistic forecasts with 9 quantiles.

  • Zero-Shot Forecasting
  • Probabilistic Forecasting
  • Covariate-Enhanced Prediction
  • High-Accuracy Time Series Analysis
  • Long-Horizon Forecasting
  • Benchmark Performance Evaluation
  • Research and Development

TimeH2O-v1 Reviews

Loading...

Popular AI Tools Like TimeH2O-v1

Chronos-2 is a 120M-parameter encoder-only time series foundation model designed for zero-shot forecasting. It supports various forecasting tasks and is efficient, delivering over…

Forecasting & Predictive Analytics
View Tool

120M-parameter encoder-only time series foundation model designed for zero-shot forecasting. It supports various forecasting tasks and is efficient, delivering over…Chronos-2 is a 120M-parameter encoder-only time series foundation model designed for zero-shot forecasting. It supports various forecasting tasks and is efficient, delivering over…

FeaturedForecasting & Predictive Analytics
See the Top 5 AI Models

Timestral-209M is a time series foundation model designed for zero-shot forecasting. It supports univariate and multivariate forecasting, probabilistic predictions, and can handle…

Forecasting & Predictive Analytics
View Tool

Time series foundation model designed for zero-shot forecasting. It supports univariate and multivariate forecasting, probabilistic predictions, and can handle…Timestral-209M is a time series foundation model designed for zero-shot forecasting. It supports univariate and multivariate forecasting, probabilistic predictions, and can handle…

Forecasting & Predictive Analytics
See the Top 5 AI Models

TimesNet

AI Models

TimesNet is a foundation model for general time series analysis, transforming 1D series into 2D space for enhanced modeling. It achieves state-of-the-art results in forecasting,…

AI Models & LLMs
View Tool

AI Models

Foundation model for general time series analysis, transforming 1D series into 2D space for enhanced modeling. It achieves state-of-the-art results in forecasting,…TimesNet is a foundation model for general time series analysis, transforming 1D series into 2D space for enhanced modeling. It achieves state-of-the-art results in forecasting,…

AI Models & LLMs
See the Top 5 AI Models

TimesFM 3.0 is a pretrained time-series foundation model developed by Google Research. It is designed for time-series forecasting and is available with PyTorch weights and…

Forecasting & Predictive Analytics
View Tool

Pretrained time-series foundation model developed by Google Research. It is designed for time-series forecasting and is available with PyTorch weights and…TimesFM 3.0 is a pretrained time-series foundation model developed by Google Research. It is designed for time-series forecasting and is available with PyTorch weights and…

Forecasting & Predictive Analytics
See the Top 5 AI Models

PatchTST

AI Models

PatchTST is an official implementation of a Transformer-based model for long-term time series forecasting. It segments time series into patches and processes them independently,…

AI Models & LLMs
View Tool

AI Models

Official implementation of a Transformer-based model for long-term time series forecasting. It segments time series into patches and processes them independently,…PatchTST is an official implementation of a Transformer-based model for long-term time series forecasting. It segments time series into patches and processes them independently,…

AI Models & LLMs
See the Top 5 AI Models

Timestral

AI GitHub Repos

Timestral is a time-series foundation model developed by Mistral AI. It is designed to handle complex time-series data, providing robust analysis and insights. Ideal for…

AI Models & LLMs
View Tool

AI GitHub Repos

Time-series foundation model developed by Mistral AI. It is designed to handle complex time-series data, providing robust analysis and insights. Ideal for…Timestral is a time-series foundation model developed by Mistral AI. It is designed to handle complex time-series data, providing robust analysis and insights. Ideal for…

AI Models & LLMs
See the Top 250 AI GitHub Repos

GluonTS

AI Frameworks

GluonTS is a Python package for probabilistic time series modeling, leveraging deep learning models built on PyTorch and MXNet. It facilitates accurate forecasting by generating…

AI Models & LLMs
View Tool

AI Frameworks

Python package for probabilistic time series modeling, leveraging deep learning models built on PyTorch and MXNet. It facilitates accurate forecasting by generating…GluonTS is a Python package for probabilistic time series modeling, leveraging deep learning models built on PyTorch and MXNet. It facilitates accurate forecasting by generating…

AI Models & LLMs
See the Top 50 AI Frameworks

Chronos

AI GitHub Repos

Chronos is a GitHub repository offering pretrained models for time series forecasting. Developed by Amazon Science, it provides tools for accurate predictions in various domains,…

Forecasting & Predictive Analytics
View Tool

AI GitHub Repos

GitHub repository offering pretrained models for time series forecasting. Developed by Amazon Science, it provides tools for accurate predictions in various domains,…Chronos is a GitHub repository offering pretrained models for time series forecasting. Developed by Amazon Science, it provides tools for accurate predictions in various domains,…

Forecasting & Predictive Analytics
See the Top 250 AI GitHub Repos
1 of 8