Models & Labs
llama.cpp b10545 Release Fixes Tensor API Bug
llama.cpp Releaseshigh confidence
Why it matters
- →Fixes a critical bug in tensor operations, preventing data corruption.
- →Enhances reliability for developers using the Tensor API.
- →Ensures accurate matrix computations even with unaligned K inputs.
The llama.cpp b10545 release resolves a significant issue in the Tensor API's mat-mat kernel, which previously caused undefined behavior when the K dimension wasn't a multiple of 32. By implementing a dynamic extent for K, the update ensures that the kernel only processes valid data, preventing potential corruption or NaN results. This change is particularly important for developers working with unaligned K inputs, as it improves the accuracy and reliability of tensor operations.
Read originalMore from llama.cpp Releases
Models & Labsmodels
llama.cpp b10541 Release Adds New Features
The b10541 release of llama.cpp enhances developer flexibility with the introduction of the --mmproj-device argument, allowing for more nuanced control over device backends. This update also maintains compatibility with existing setups through the MTMD_BACKEND_DEVICE environment variable and introduces a convenient -mmdev shortflag. These improvements make it easier for developers to manage and load device backends efficiently. While there are no new model architectures in this release, the focus on refining usability ensures that developers can deploy their applications smoothly across different environments.
llama.cpp Releases
Open Sourcemodels
llama.cpp b10566 Release Expands Platform Support
The b10566 release of llama.cpp focuses on broadening its reach across various architectures, though it doesn't bring any groundbreaking innovations. This update includes support for macOS, Linux, Windows, and openEuler, with some configurations like macOS Apple Silicon with KleidiAI and Ubuntu x64 with ROCm 7.14 being disabled. The release ensures that llama.cpp remains a versatile tool, offering Vulkan and OpenVINO support on different systems. While it doesn't introduce new features, it solidifies llama.cpp's role as a flexible inference runtime, accommodating a wide array of hardware environments.
llama.cpp Releases
Open Sourcemodels
llama.cpp b10568 Release Expands Platform Support
The latest b10568 release of llama.cpp continues its trend of broadening platform compatibility, now incorporating the ggml_rope_set_offset() function. This update partially applies to deepseek2, enhancing its functionality. The release maintains support across a wide array of systems, including macOS, Linux, Windows, and openEuler, with specific configurations for Vulkan, ROCm, and CUDA environments. While no groundbreaking features are introduced, this update solidifies llama.cpp's position as a versatile tool for developers working across diverse hardware setups.
llama.cpp Releases
More in Models & Labs
Models & Labsmodels
v0.28.0rc2: DFlash2 Local Convolution Update
The v0.28.0rc2 release of vLLM introduces DFlash2, a feature that enhances local convolution capabilities with a candidate selector. This update, cherry-picked from a specific commit, signifies a technical refinement aimed at improving model performance. While the specifics of the implementation are technical, the focus on local convolution suggests a targeted improvement in processing efficiency. This release is a step forward for developers looking to optimize their AI models with more precise convolution operations.
vLLM Releases
© Lev SelectorModels & Labsmodels
Qwen3.8-27B Model Released with MTP for Faster Performance
The Qwen3.8-27B model has been released, featuring MTP technology for enhanced speed and local operation capabilities.
Lev Selector
© Lev SelectorModels & Labsmodels
GLM-5.3 Outperforms Mythos 5 in Cybersecurity
The open-source model GLM-5.3 has surpassed Mythos 5 in cybersecurity performance, showcasing advancements in Chinese AI models.
Lev Selector