Topic: attestations

38 stories found

Yesterday

releases2 sources⚡ Corroborated48

ggml/llama.cpp releases: b11118

The ggml/llama.cpp project released a new version that includes an update to introduce a direct-mapped DMA cache for better handling of HVX FA mask operations, enhancing performance. This update is significant as it optimizes memory management, particularly relevant for Apple Silicon (arm64) systems on macOS and iOS.

Covered by ggml/llama.cpp releases

Tuesday, September 22, 2026

releases2 sources⚡ Corroborated48

ggml/llama.cpp releases: b11094

The ggml/llama.cpp project updated the cpp-httplib library to version 0.57.1, a change signed off by Adrien Gallouët that enhances the project's functionality. This update is significant as it improves compatibility and performance for users of the llamacpp software suite.

Covered by ggml/llama.cpp releases

releases48

ggml/llama.cpp releases: b11115

The ggml/llama.cpp project has released a new version including an OpenCL bin kernel for the `kernel_gemm_noshuffle_q4_k_q8_1_dp4a_ila_a8_bin`, enhancing binary kernel selection and support for non-MoE dp4a operations, which is crucial for optimizing large language model computations on specific hardware.

github.com↗

Sunday, September 20, 2026

releases2 sources⚡ Corroborated48

ggml/llama.cpp releases: b11065

The ggml/llama.cpp project released an update that tunes the FA parameter for use with the Gemma 4 on Ampere or newer GPUs, enhancing performance. This update is significant as it optimizes machine learning model processing for specific hardware, potentially improving speed and efficiency in applications like natural language processing.

Covered by ggml/llama.cpp releases

releases48

ggml/llama.cpp releases: b11063

The ggml/llama.cpp project released a new version addressing issues with handling invalid UTF-8 sequences in the Abstract Syntax Tree (AST), improving Unicode compliance and flexibility. This update is crucial for enhancing the robustness of text processing functionalities, ensuring better compatibility and reliability.

github.com↗

Saturday, September 19, 2026

releases2 sources⚡ Corroborated48

ggml/llama.cpp releases: b11056

The ggml/llama.cpp project released version b11056, which includes a change to enable I32 GET_ROWS (#29116). This update is significant for improving the handling of 32-bit integer operations in the library, potentially enhancing performance and functionality.

Covered by ggml/llama.cpp releases

Friday, September 18, 2026

releases3 sources🛡️ Verified48

ggml/llama.cpp releases: b11046

The ggml/llama.cpp project has released an update that adds support for the `flash_attn_f32_f16_bin` kernel in OpenCL, enhancing computational efficiency for certain operations. This update is significant as it improves performance in processing tasks related to large language models.

Covered by ggml/llama.cpp releases

Thursday, September 17, 2026

releases4 sources🛡️ Verified48

ggml/llama.cpp releases: b11028

The ggml/llama.cpp project released version b11028, which includes a fix for evicting old files. This update is important as it enhances the project's stability and efficiency, particularly for users on Apple Silicon platforms.

Covered by ggml/llama.cpp releases

releases2 sources⚡ Corroborated48

ggml/llama.cpp releases: b11011

The ggml/llama.cpp project released a fix for the function signature of `ggml_backend_sycl_split_buffer_type` to address compatibility issues, which is important for ensuring smooth operation on specific hardware platforms. This update affects users particularly interested in Apple Silicon Macs and iOS devices.

Covered by ggml/llama.cpp releases

Wednesday, September 16, 2026

releases48

ggml/llama.cpp releases: b11007

The ggml/llama.cpp project released a new version that enables CUDA graph usage for MTP, improving performance. This update is significant as it enhances the efficiency of the software, particularly relevant for users requiring high computational power.

github.com↗

38 of 38 items shown. Sources: 124 days indexed.