llamafile
llamafile, Mozilla's project to package an LLM as a single, executable cross-platform binary. Run local AI with one file.
26k
Stars
1.6k
Forks
4
Releases tracked
GitHub releases
Source type
22 days
Avg cadence
May 2026
First tracked
24d ago
Latest
Release history
llamafile 0.10.5
Focused on simplifying the upstream llama.cpp sync task
llamafile 0.10.4
Introduces first version of transcribefile
llamafile 0.10.3
Fixes SIGSEGV regression in GPU initialization, restoring CPU fallback when GPU setup fails
llamafile 0.10.2
Llamafile 0.10.2 improves GPU acceleration detection, fixes CPU flash-attention and MoE matmul operations, and reduces CUDA library size