PulseAugur
EN
LIVE 20:23:39
ENTITY ServeurpersoCom

ServeurpersoCom

PulseAugur coverage of ServeurpersoCom — every cluster mentioning ServeurpersoCom across labs, papers, and developer communities, ranked by signal.

Show in brief
Total · 30d
5
10 over 90d
Releases · 30d
0
0 over 90d
Papers · 30d
0
0 over 90d
TIER MIX · 90D
TOPICS
SENTIMENT · 30D

3 day(s) with sentiment data

RECENT · PAGE 1/1 · 10 TOTAL
  1. TOOL · CL_362262 ·

    llama.cpp b12461 fixes Vulkan exception reporting on Apple

    The llama.cpp project has released version b12461, which includes a fix for tests related to Vulkan on Apple devices. This update ensures that exception messages are printed, addressing intermittent failures that were p…

  2. TOOL · CL_353520 ·

    llama.cpp update fixes long context overflow issue

    The llama.cpp project has released an update, b12293, which addresses an overflow issue in its CPU kernel for handling long contexts. The update modifies the kernel to accumulate V cache in F32 format, preventing overfl…

  3. TOOL · CL_342956 ·

    llama.cpp b12128 release fixes TOPK_MOE CI flakes

    The llama.cpp project released version b12128, which includes fixes for CI flakes related to TOPK_MOE. These flakes were caused by near ties in the single-precision floating-point format when using the sqrt_softplus gat…

  4. TOOL · CL_301887 ·

    llama.cpp updates KV cache handling for partial state serialization

    The llama.cpp project has released an update, tagged b11644, which refines how the key-value (KV) cache handles partial state serialization. This update ensures that the cache correctly honors the PARTIAL_ONLY flag, pre…

  5. TOOL · CL_300093 ·

    llama.cpp b11626 release fixes layer splitting bug

    The llama.cpp project has released version b11626, which addresses an issue where layers were not split evenly when no device reported free memory. Previously, a NaN split caused layers to be sent to an invalid device i…

  6. TOOL · CL_283740 ·

    llama.cpp removes GLM5-Next sparse attention gather path

    The llama.cpp project has released an update, version b11453, which removes the gather path for GLM5-Next sparse attention. This change simplifies the attention mechanism by relying on a plain matmul and softmax, with f…

  7. TOOL · CL_281357 ·

    llama.cpp b11422 optimizes CUDA/ROCm with new MUSA indexer kernel

    The llama.cpp project has released version b11422, which includes optimizations for CUDA and ROCm architectures. Specifically, the update introduces a new vector lightning indexer kernel for MUSA, addressing shared memo…

  8. TOOL · CL_278408 ·

    Qwen4Exp optimization reduces indexer score memory in llama.cpp

    A pull request has been submitted to the llama.cpp project to optimize the Qwen4Exp model. This optimization aims to reduce the memory required by the indexer score, potentially allowing the Qwen Flash Next model to use…

  9. TOOL · CL_136931 ·

    llama.cpp server accepts null sampling parameters, aligning with OpenAI spec

    The llama.cpp project has released version b9967, introducing an update to its server functionality. This update allows the server to accept null sampling parameters, such as temperature and top_p. By treating null valu…

  10. TOOL · CL_61760 ·

    llama.cpp adds custom CSS theming to its web UI

    The llama.cpp project has released version b9438, introducing a new feature that allows for custom CSS injection into its web UI. This update enables users to theme the interface by providing custom CSS through a config…