github llama.cpp · 3d ago · inference

llama.cpp b10934 implements common_schema for JSON schemas

from English

The llama.cpp project released build b10934, which introduces the implementation of a `common_schema` internal representation for JSON schemas. This update includes the creation of `common_schema` types and a JSON schema optimizer to handle schema resolution more effectively.

Implementation of `common_schema` types and type/kind resolution logic.
Introduction of a JSON schema optimizer and refactoring of json-schema-to-grammar to use the new common schema structure.
Renaming of `common_schema_parse` to `common_schema_from_json` and builder methods from `parse_*` to `build_*`.
Handling of empty tool parameters, empty `$ref`, and ensuring `{}` means any object for `json_object`.
Restoration of `(min|max)Length` to imply string type and moving enums under `common_schema`.

The release provides binaries for macOS (Apple Silicon and Intel), Linux (Ubuntu x64, arm64, s390x with CPU, Vulkan, ROCm, OpenVINO, SYCL), Windows (CPU, CUDA 12/13, Vulkan, OpenCL, OpenVINO, SYCL, ROCm), Android, and iOS XCFrameworks.

Importance 1/3 Trust 1/3 llama.cpp Inference efficiency

Related articles

github llama.cpp · 1h ago · 1 view Live

llama.cpp b10984 adds CUDA row-contiguous SUM_ROWS support

The llama.cpp project released version b10984, introducing CUDA support for row-contiguous SUM_ROWS operations. This update also organizes the code to include GGML_OP_MEAN for handling row-contiguous tensors via a shared kernel.

github llama.cpp · 2h ago · 1 view

llama.cpp b10985 restricts RPC hash-cache to weights

The llama.cpp project released build b10985, which modifies the RPC backend to limit hash-based caching exclusively to weight tensors. This change addresses an issue where activation data was unnecessarily hashed and stored, leading to excessive disk usage.

github llama.cpp · 3h ago · 4 views

llama.cpp b10982 adds Vulkan sparse Flash Attention support for DSV4/GLM

The llama.cpp project has released build b10982, which introduces support for sparse Flash Attention via the Vulkan backend. This update specifically targets DSV4 and GLM models, allowing them to leverage sparse attention mechanisms on compatible GPU hardware.

github llama.cpp · 4h ago · 2 views

llama.cpp b10981 fixes stateful decode for Gemma-4 and optimizes GPU MoE inference

The llama.cpp b10981 release addresses critical correctness issues in the OpenVINO backend's stateful decoding path, specifically fixing support for models like Gemma-4 that use per-layer head sizes. It also introduces optimizations for GPU Mixture of Experts (MoE) inference and improves performance through KV state relayouts.

github llama.cpp · 5h ago · 1 view

llama.cpp adds Metal FA kernels for MiniCPM3's HSK=96, HSV=64

The llama.cpp project has added Flash Attention (FA) kernel support for the specific configuration used by MiniCPM3 on Apple Silicon. This change addresses a crash where the `-fa auto` flag would abort due to a missing kernel instantiation for attention key length 96 and value length 64.