llama.cpp Releases
609 releases of ggml-org/llama.cpp
- b10531
Reverts a previous fix for the tensor-split meta backend to address issues introduced by the change, restoring the previous behavior.
Aug 21, 2026
- b10520
Fixes a compilation issue on 32-bit ARM architectures by gating the use of `__fp16` on the IEEE half format, ensuring compatibility with toolchains that provide the type but use a different format.
Aug 20, 2026
- b10519
Refactors the server's sleep handling logic to allow access to the `/metrics` endpoint even when the server is in a sleeping state, improving observability and fixing race conditions.
Aug 20, 2026
- b10517
Optimizes Flash Attention in the Vulkan backend by dequantizing KV cache only once in cooperative matrix operations and handling edge cases like FA scratch buffer limits.
Aug 20, 2026
- b10516
Addresses a potential crash in the Vulkan backend by adding null checks in the queue command pools cleanup function.
Aug 20, 2026
- b10514
Introduces support for new Granite model architectures featuring Sliding Window Attention (SWA), including GraniteSWAForCausalLM and GraniteMoeSWAForCausalLM, along with infrastructure for converting rope patterns.
Aug 20, 2026
- b10509
Adds a new parameter `ggml_rope_set_offset` to control Rotary Positional Embedding (RoPE) offsets, with full support implemented for CPU, Metal, CUDA, and Vulkan backends.
Aug 20, 2026
- b10507
This release introduces a minor update to the Multi-Modal (mtmd) module, specifically adding functionality for bitmap set merging to improve memory management.
Aug 20, 2026
- b10506
Optimizes the macOS Metal backend by implementing q8_0 dequantization using packed types, which should improve performance on Apple Silicon.
Aug 20, 2026
- b10505
Introduces a new server preset option to deduplicate cached models, allowing for more efficient memory usage and cache management.
Aug 20, 2026
- b10502
Adds CI attestation capabilities to verify the integrity and origin of signed release artifacts, enhancing security and trust in the build process.
Aug 19, 2026
- b10488
Upgrades the OpenVINO backend to version 2026.3 and updates device drivers, while skipping a specific CI rollback test for the Nemotron-H model.
Aug 18, 2026
- b10486
Resolves a bug in the multi-modal model (mtmd) where the LFM2 image tiling threshold was incorrectly processed, ensuring compatibility across platforms.
Aug 18, 2026
- b10485
Synchronizes core ggml library code with the main repository, incorporating necessary updates from the upstream source.
Aug 18, 2026
- b10483
Improves build system robustness by fixing xcframework generation, removing unused include paths, and standardizing vendored library targets using CMake aliases.
Aug 18, 2026
- b10472
Fixes memory reporting on AMD APUs within HIP builds to prevent over-promising available memory on systems with small memory carve-outs.
Aug 17, 2026
- b10470
This release updates the CI/CD pipeline to explicitly push the release tag using git tag before creating the release via the GitHub API, ensuring the tag is available for workflows.
Aug 17, 2026
- b10456
This release fixes thread and block count calculations in quantized copy kernel launches for the SYCL backend, significantly improving throughput for certain quantization types.
Aug 17, 2026
- b10455
This release adds support for AdamW and SGD optimizer steps in the SYCL backend, along with documentation updates.
Aug 17, 2026
- b10453
This release removes some instances of the ggml_concat operation as part of refactoring efforts.
Aug 16, 2026