Search results

Follow-up to #978: extend fa_helpers multi-arch coverage beyond AVX-512F

Status: Open.

#981 In mozilla-ai/llamafile;

· aittalam opened on May 24, 2026

Follow-up to #978: dense-model CPU performance gap vs upstream llama.cpp

Status: Open.

#980 In mozilla-ai/llamafile;

· aittalam opened on May 24, 2026

🎉 Your project is featured in awesome-ai-chat-apps

Status: Open.

#979 In mozilla-ai/llamafile;

· yunshan25 opened on May 24, 2026

Bug: CPU performance regression

bug

medium severity

Status: Open.

#973 In mozilla-ai/llamafile;

· Roysten opened on May 20, 2026

Bug: --server --host <unix-socket-path> silently fails to bind on Linux (v0.10.1)

Status: Open.

#971 In mozilla-ai/llamafile;

· 3rg0n opened on May 18, 2026

Bug: lalmafile save queries to file even --disable-log is specified

bug

medium severity

Status: Open.

#966 In mozilla-ai/llamafile;

· ghost opened on May 14, 2026

Bug: No usable GPU found, --gpu-layers option will be ignored but AMD GPU does exist

bug

medium severity

Status: Open.

#965 In mozilla-ai/llamafile;

· ghost opened on May 14, 2026

Bug: Not clear how to use it with Cuda on Windows (10)

bug

medium severity

Status: Open.

#961 In mozilla-ai/llamafile;

· nor-bee opened on May 6, 2026

Bug: Please support the audio analysis capability of gemma-4-E4B-it-GGUF

bug

medium severity

Status: Open.

#960 In mozilla-ai/llamafile;

· Leonard-Li777 opened on May 5, 2026

Integration: PaperClaw (autonomous peer-reviewed paper generator for Llamafile)

Status: Open.

#947 In mozilla-ai/llamafile;

· Agnuxo1 opened on Apr 16, 2026

Bug: Vulkan support on Windows does not work consistently

bug

high severity

Status: Open.

#938 In mozilla-ai/llamafile;

· aittalam opened on Apr 8, 2026

Improve Vulkan dylib speed

enhancement

Status: Open.

#931 In mozilla-ai/llamafile;

· aittalam opened on Mar 31, 2026