# Keras v3.15.0

- Product: Keras (https://whatsnew.fyi/product/keras)
- Vendor: Keras
- Date: 2026-06-24
- Version: v3.15.0
- Original notes: https://github.com/keras-team/keras/releases/tag/v3.15.0
- Permalink: https://whatsnew.fyi/product/keras/releases/v3.15.0

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **added** — Added `export_torch` function to export Keras models to native PyTorch `nn.Module` format
- **added** — Added `sliding_window` parameter to `MultiHeadAttention` and `GroupedQueryAttention` for efficient long-context attention
- **added** — Added `unique`, `fabs`, `fmax`, `fmin`, `dsplit`, `erfc`, `percentile`, and `nanpercentile` operations to `keras.ops.numpy`
- **added** — Added `pinv` (pseudo-inverse) and `matrix_rank` operations to `keras.ops.linalg`
- **added** — Added `sobel_edges` and `ssim` (structural similarity) operations to `keras.ops.image`
- **added** — Added `MultiOptimizer` to support training sub-networks with different optimizers
- **added** — Added `predict_proba` method to `SKLearnClassifier`
- **added** — Added LiteRT (TFLite) export support for models using the PyTorch backend
- **added** — Added CTC beam search decoding for the Torch backend
- **changed** — Causal-only attention in `MultiHeadAttention` and `GroupedQueryAttention` now automatically dispatches to Flash Attention (cuDNN SDPA)
- **changed** — `keras.ops.transpose` now supports negative axis values
- **changed** — JAX backend now fuses Bidirectional LSTM into a single cuDNN call
- **changed** — Added fused bidirectional GRU for Torch backend
- **changed** — Added distribution information (`num_processes`, `num_model_replicas`, `data_shard_id`) to `ModelParallel`
- **fixed** — Fixed manual attention path to correctly apply causal masking
- **fixed** — Fixed LiteRT export for Keras 3 with TensorFlow 2.20 and Python 3.13
- **fixed** — Fixed float16 numerical instability in `GroupNormalization` with small epsilon
- **fixed** — Fixed GPU OOM with rank-3 input in Dense layer due to `BatchMatMulV2` gradient materialization
- **security** — Hardened model reloading against HDF5 exploits by rejecting `ExternalLink`, `SoftLink` groups, virtual datasets, and shape-bomb datasets
- **security** — Hardened archive loading by rejecting tar members and links escaping extraction directory and ZIP/NPZ members declaring excessive data

##### Highlights

- **Keras-to-Torch Export**: New `export_torch` enables exporting Keras models to native PyTorch `nn.Module` format, along with LiteRT (TFLite) export support for the PyTorch backend.
- **Sliding Window Attention**: Added `sliding_window` parameter to `MultiHeadAttention` and `GroupedQueryAttention` for efficient long-context attention.
- **Flash / Fused SDPA**: Causal-only MHA/GQA now automatically dispatches to Flash Attention (cuDNN SDPA), and the manual attention path correctly applies causal masking.
- **Multi-Optimizer Training**: New `MultiOptimizer` supports assigning different optimizers to sub-networks.
- **New Math Operations**: Added `unique`, `pinv`, `matrix_rank`, `fabs`, `fmax`, `fmin`, `erfc`, `dsplit`, `percentile`, `nanpercentile`, `sobel_edges`, and `ssim` (structural similarity) to `keras.ops`.
- **Security Hardening**: Comprehensive hardening of model reloading against HDF5 exploits, tar/zip traversal attacks, insecure deserialization.

---

##### New Features and Operations

###### Multi-Backend Operations
- **New NumPy Operations**: Added `unique`, `fabs`, `fmax`, `fmin`, `dsplit`, `erfc`, `percentile`, `nanpercentile` in `keras.ops.numpy`.
- **New Linear Algebra Operations**: Added `pinv` (pseudo-inverse) and `matrix_rank` in `keras.ops.linalg`.
- **New Image Operations**: Added `sobel_edges` for edge detection and `ssim` (structural similarity) in `keras.ops.image`.
- **Negative Axes in Transpose**: `keras.ops.transpose` now supports negative axis values.

###### Layers and Attention
- **Sliding Window Attention**: `MultiHeadAttention` and `GroupedQueryAttention` layers support the `sliding_window` parameter for efficient long-sequence processing.
- **Flash Attention Engagement**: Causal-only attention in MHA/GQA now uses Flash SDPA for significant speedups.
- **Fused Bidirectional LSTM/GRU**: JAX backend now fuses Bidirectional LSTM into a single cuDNN call; fused bidirectional GRU added for Torch backend.
- **CTC Beam Search Decoder**: Added CTC beam search decoding for the Torch backend.

###### Training and Optimizers
- **MultiOptimizer**: Supports training sub-networks with different optimizers.
- **SKLearn Classifier**: Added `predict_proba` method to `SKLearnClassifier`.

---

##### Export and Deployment

- **Keras-to-Torch Export**: Export Keras models to native PyTorch `nn.Module` via `model.export(..., format="torch")`.
- **LiteRT (TFLite) Export for PyTorch**: Added LiteRT export support for models using the PyTorch backend.
- **LiteRT Compatibility Fix**: Fixed LiteRT export for Keras 3 + TF 2.20 + Python 3.13.
- **ONNX Export**: Support for dict/list inputs in Torch ONNX export; documented static input signature requirement for LiteRT PyTorch export.

---

##### Distribution and Parallelism

- **ModelParallel Improvements**: Defined contiguous replica-group data shard ID convention; added distribution information (`num_processes`, `num_model_replicas`, `data_shard_id`).
- **Initializer Distribution Layout**: Initializers can now handle the distribution layout directly with JAX.
- **TF Dataset Distribution**: Refactored TF dataset distribution with centralized sharding routing; fixed data distribution for model training in JAX.

---

##### OpenVINO Backend Support

The OpenVINO backend received continued improvements:

- **New Operations**: Implemented `glu`, `sparsemax`, `gaussian_blur`, `logdet`, `cholesky`, `lu_factor`, `erfc`, `segment_min`, `segment_prod`, `percentile`, `nanmedian`, `nanpercentile`, `unique`, `flash_attn`, `greedy ctc_decode`, `solve_triangular`, `compute_homography_matrix`, and image transforms (affine, perspective, elastic).
- **Opset Upgrades**: Upgraded to opset16 for select operations and full upgrade.
- **Fixes**: Dynamic/symbolic shape handling, mask propagation, random seed determinism, dropout during predict, Lanczos interpolation in resize, dynamic batch shape propagation, and improved ef

_[Truncated at 4000 characters — full notes: https://github.com/keras-team/keras/releases/tag/v3.15.0]_