llama.cpp Releases
684 releases of ggml-org/llama.cpp
- b10058
Adds support for Q2_0 quantization format to the Vulkan backend and optimizes mat-vec-mul performance.
Jul 17, 2026
- b10057
Fixes critical bugs in the SYCL backend related to quantization row calculations and Q5_K kernel processing.
Jul 17, 2026
- b10056
Adds an OpenCL operation for absolute value (ABS) to the backend.
Jul 17, 2026
- v0.7.0
This release is for the WeKnora project (not llama.cpp), introducing QQBot integration, MCP OAuth features, and extensive security fixes.
Jul 17, 2026
- b10054
Adds documentation notes regarding OpenCL usage with specific Adreno 810 hardware and includes standard release binaries.
Jul 17, 2026
- b10052
Major overhaul of Hexagon L2 cache handling, introducing dirty bit tracking, threaded flushing, and various MUL_MAT and workqueue improvements.
Jul 16, 2026
- b10051
Updates KleidiAI integration to distinguish between SME (v1) and SME2 hardware capabilities, ensuring correct kernel dispatch.
Jul 16, 2026
- b10050
Fixes a potential race condition in Vulkan by synchronizing on event wait when using transfer queues for asynchronous copies.
Jul 16, 2026
- b10048
This release fixes Transformer Parallelism (TP) support for specific model architectures including Phi3, Bert, Plamo2/3, and ChatGLM.
Jul 16, 2026
- b10047
Updates the BoringSSL vendor dependency to version 0.20260713.0 and provides pre-built binaries for macOS, Linux, Windows, and Android.
Jul 16, 2026
- b10046
Improves test coverage by implementing the `test-recurrent-state-rollback` test case and includes pre-compiled binaries for all supported platforms.
Jul 16, 2026
- b10045
Adds a server feature allowing text-only slot save and restore when using Multimodal (mtmd) models, accompanied by standard platform binaries.
Jul 16, 2026
- b10043
Adds support for CUDA virtual devices, allowing the simulation of multiple GPUs, and disables NCCL in this scenario.
Jul 16, 2026
- b10042
Activates CUDA graph support for Volta and Turing architecture GPUs to improve performance.
Jul 16, 2026
- b10041
Reduces log noise in the server by ignoring empty or missing Origin headers in the CORS handling.
Jul 16, 2026
- b10040
Fixes a bug in CUDA/HIP integration detection where the integrated flag was incorrectly set to false, preventing host buffer usage on AMD APU parts.
Jul 16, 2026
- b10038
Updates release notes to include a link to the official website.
Jul 16, 2026
- b10037
Enhances the quantization process to support manual tensor types when using the `--pure` flag.
Jul 16, 2026
- b10036
Disables specific OpenCL kernels (FA and MoE) on Adreno 850 to prevent compiler-related errors, and fixes a parsing bug for DirectX compiler versions.
Jul 16, 2026
- b10035
Optimizes the CUDA kernel for Q1_0 quantization type using the `__byte_perm` intrinsic instruction.
Jul 16, 2026