Skip to content

LLM inference in C/C++

C++ ggml

Commits on master

cuda: extract Q1_0 elements via __byte_perm (#25628)
David Friehs committed 4 months ago
602f828
opencl: exclude some moe kernels on Adreno a7x (#25698)
Hongqiang Wang committed 4 months ago
505b1ed
ui: Agentic Content UX improvements (#25450)
Aleksander Grygier committed 4 months ago
32beb24
tokenize : drop --stdin mutual-exclusion check (#25672)
Adrien Gallouët committed 4 months ago
aff6eb6
c3d47e6
f6f12e4
Fix crash with draft-simple (#25720)
Gaurav Garg committed 4 months ago
956973c
ci : add HF_TOKEN to self-hosted workflows (#25706)
Georgi Gerganov committed 4 months ago
c810293
DeepseekV4: reduce graph splits (#25702)
Aman Gupta committed 4 months ago
33a75f4
sycl : fix get_rows Q2_K, Q4_K, Q5_K (#25656)
Neo Zhang committed 4 months ago
d3fba0c
ae9291e
sycl : implement xielu op (#25550)
Andrew Smith committed 4 months ago
22b208b
server : refactor prompt cache state ownership (#25649)
Georgi Gerganov committed 4 months ago
bf2c86d
server: add --cors-* options (#25655)
Xuan-Son Nguyen committed 4 months ago
6e52db5
ui: Fix spacing in tool-call request (#25634)
Bill Sideris committed 4 months ago
236ab57