Skip to content

LLM inference in C/C++

C++ ggml

Commits on master

metal : add sparse FA (#28098)
Georgi Gerganov committed 2 months ago
7bb0fc1
metal : fix glu dispatch with ne00 = 1 (#28306)
Georgi Gerganov committed 2 months ago
0df017d
mtmd : add `const` in various places (#28307)
Mads Marquart committed 2 months ago
f45576a
0ba6499
vulkan: fix FA dequant path engagement (#28190)
Nathan Wilson committed 2 months ago
c7bda03
5ec4eab
mtmd: fix idefics3 preproc (#28273)
Xuan-Son Nguyen committed 2 months ago
67a17c1
finetune: fix no KV cache (#27199)
Xuan-Son Nguyen committed 2 months ago
159b741
f027c4f
mtmd: add mtmd_tokenize_from_parts() (#28250)
Xuan-Son Nguyen committed 2 months ago
7339054
metal : add fa-vec tunings for M3 (#28236)
Isaac committed 2 months ago
9cc3394
8c0b9cd
ci : check for missing autoreleasepools (#27884)
Niklas Wenzel committed 2 months ago
03dbcc5
Update ROCm to 10.0.0 release (#27803)
Mario Limonciello committed 2 months ago
cff1844
9400c89
ci : enable hf-jobs on server-cuda (#28258)
Sigbjørn Skjæret committed 2 months ago
d5fec32
ggml-cuda : remove unused vars (#28235)
Adrien Gallouët committed 2 months ago
3d3d7c8
mtmd: support DeepSeek-V4-Flash-Vision-Exp (#28133)
Xuan-Son Nguyen committed 2 months ago
7798007
CUDA + ggml: add sparse-fa for DSV4/GLM (#27970)
Aman Gupta committed 2 months ago
8e93a97
mtmd: Fix Qwen3-tts-0.6b (#28231)
Pascal committed 2 months ago
0f3a71b
ggml: avoid KleidiAI buffer type init on dispatch (#27891)
Aman Chadha(IVIXMMI) committed 2 months ago
b81c99b
hexagon: MUL_MAT and MUL_MAT_ID fusion and fixes (#28202)
Max Krasnyansky committed 2 months ago
960dffa