llamafile · Issues· 211 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #980
Follow-up to #978: dense-model CPU performance gap vs upstream llama.cpp
Updated Sep 11, 2026 - #1048
Bug: --gpu nvidia and vulkan fail on older NVIDIA mobile GPUs (GTX 970M / Maxwell) on Windows
bughigh severityUpdated Sep 5, 2026 - #1047
Bug: `/stats` command prints nothing in TUI chat
buglow severityUpdated Sep 4, 2026 - #931
Improve Vulkan dylib speed
enhancementUpdated Aug 11, 2026 - #803
Feature Request: Mobile support android / ios
enhancementUpdated Jul 26, 2026 - #938
Bug: Vulkan support on Windows does not work consistently
bughigh severityUpdated Jul 24, 2026 - #438
Is it possible for llamafile to use Vulkan or OpenCL Acceleration?
request to lend supportUpdated Jul 24, 2026 - #226
GPU speed-up on Raspberry Pi 5
request to lend supportvulkanUpdated Jul 24, 2026 - #806
Bug: Cannot run llamafile on OpenBSD: syntax error: NUL byte unexpected
bugcritical severityUpdated Jul 16, 2026 - #1017
Sandbox combined mode (TUI + server) instead of skipping entirely
enhancementUpdated Jul 14, 2026 - #1012
Feature Request: Update vendored mbedtls
enhancementUpdated Jul 2, 2026 - #35
Homebrew formula
request to lend supportUpdated Jun 24, 2026 - #728
Bug: Error when parsing wav file that ffmpeg understands
bugmedium severityUpdated Jun 6, 2026 - #15
Security
questionUpdated May 28, 2026 - #981
Follow-up to #978: extend fa_helpers multi-arch coverage beyond AVX-512F
Updated May 24, 2026