Topic: cuda

The ggml/llama.cpp project fixed a compilation error related to CUDA sm_70 tiles by generalizing the tile shape in version b11090, addressing an issue where a recent update mismatched tile definitions. This update is crucial for ensuring compatibility across different GPU architectures.

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.

The ggml/llama.cpp project updated its Windows ARM64 CUDA 13.4 builds to use the General Availability (GA) redistributables of version 13.4.1, moving away from Developer Preview archives. This update is important as it ensures compatibility and stability for users running the software on compatible hardware.

vLLM released version 0.29.0, which includes 594 commits from 277 contributors and marks the full rollout of Model Runner V2 as the default for all models, enhancing performance with CUDA graph memory profiling for KV cache auto-sizing. This update is significant as it improves model efficiency and scalability in large language model deployments.

The ggml/llama.cpp project reverted a commit to restore `prop.integrated` on HIP builds, addressing potential issues with CUDA and HIP compatibility. This change is important for maintaining the library's functionality across different hardware platforms.