github vLLM · 2d ago · inference

vLLM v0.29.0 applies dense prefix cache default to hybrid models

from English

The vLLM project released version 0.29.0, which includes a core bugfix regarding the dense prefix cache.

The update applies the dense prefix cache default setting specifically to hybrid models.

Importance 1/3 Trust 2/3 vLLM Inference efficiency

llama.cpp b10891 fixes PowerVR Vulkan crash by falling back to shared-memory reduction

The llama.cpp project released version b10891, which addresses a critical stability issue on PowerVR GPUs. The update modifies the Vulkan backend to fall back to shared-memory reduction for dequantized matrix-vector multiplication (dmmv) operations.

media MarkTechPost · just now Live

DeepSeek releases DeepSeek-V4.1-Flash with 1M context and FP4 KV cache

DeepSeek AI has released DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model featuring a 1-million-token context window and an FP4 KV cache that reduces global cache footprint to 890 bytes per token. The model utilizes a causal encoder-decoder architecture and Compressed Sparse Attention 2 (CSA2) to significantly lower memory requirements while maintaining performance.

github llama.cpp · 1h ago · 1 view Live

llama.cpp b10891 fixes PowerVR Vulkan crash by falling back to shared-memory reduction

DeepSeek releases DeepSeek-V4.1-Flash with 1M context and FP4 KV cache

llama.cpp b10889 release avoids V cache allocation for indexer

llama.cpp b10888 adds Vulkan command-buffer debug labels for GPU profilers

llama.cpp b10886 adds Q1_0 vector intrinsic support for s390x