Models & Labs

Google unveils TabFM for zero-shot tabular data prediction

Google Research BlogJune 30, 2026high confidence

Why it matters

→TabFM simplifies the traditionally complex process of tabular data prediction by eliminating manual training and tuning.
→It leverages synthetic datasets to generalize well to real-world data, addressing the scarcity of open-source tabular datasets.
→The integration into Google BigQuery democratizes access to advanced machine learning capabilities for non-experts.

Google unveils TabFM for zero-shot tabular data prediction — ©Google Research Blog

Google Research has launched TabFM, a new foundation model for tabular data that simplifies classification and regression tasks. Unlike traditional models that require extensive training and tuning, TabFM uses in-context learning to make predictions without manual intervention. This model is trained on synthetic datasets to capture complex feature interactions, outperforming existing methods in benchmarks. TabFM will soon be integrated into Google BigQuery, allowing users to perform advanced data analysis with simple SQL commands.

Read original

More from Google Research Blog

Models & Labsmodels

Google unveils SensorFM for wearable health data

Google Research has introduced SensorFM, a foundation model trained on over a trillion minutes of wearable sensor data from five million people. This model represents a significant leap in wearable health technology, offering a general-purpose representation of human physiology that can be adapted to various health prediction tasks. By leveraging self-supervised learning, SensorFM effectively handles fragmented data, a common issue with wearable devices. This development could transform wearable health research by moving away from single-outcome models to a more comprehensive approach, potentially enhancing personalized health insights.

Google Research BlogJul 9, 2026

More in Models & Labs

Models & Labsmodels

Llama.cpp b9946 Release Enhances Hexagon Optimizations

The latest b9946 release of llama.cpp focuses on optimizing Hexagon operations, particularly unary operations, to improve performance and efficiency. By introducing tiling for wide rows and replacing divisions with fastdiv, the update aims to prevent VTCM overflow and streamline code execution. The release also includes tracing instrumentation and specialized thread functions to enhance code generation. While no new models are introduced, these technical improvements make llama.cpp more robust and efficient for developers working with Hexagon architectures.

llama.cpp ReleasesJul 11, 2026

Models & Labsmodels

Llama.cpp b9948 Release Enhances CUDA Efficiency

The latest b9948 release of llama.cpp focuses on optimizing memory usage in CUDA operations, specifically in the ggml_top_k() and ggml_argsort() functions. By processing data in smaller chunks, the update reduces the need for large temporary buffers, enhancing performance on CUDA-enabled systems. This release also includes minor code improvements like allocating temporary destinations only once and refining the use of ternary operators. While no new model architectures are introduced, these changes make llama.cpp more efficient for developers working with CUDA, particularly in memory-constrained environments.

llama.cpp ReleasesJul 11, 2026

Models & Labsmodels

llama.cpp b9951 Release Enhances ET Backend

The latest b9951 release of llama.cpp marks a significant enhancement in the ET backend, introducing a range of new kernels and performance optimizations. This update includes the addition of various matrix operations and support for FlashAttention, which promises to improve computational efficiency. The release also focuses on vectorization and parallelization, aiming to boost performance across different operations. These changes make the ET backend more robust and capable, potentially benefiting developers working with complex AI models by offering improved speed and functionality.

llama.cpp ReleasesJul 11, 2026