COMMITS
June 22, 2026
E
Add Laguna M.1 GGUF support (#2003)
empty-quiver committed
J
DFlash: use persistent FA-ready K/V cache (#1997)
Joel Farthing committed
M
on-demand tensor reload (#1989)
magikRUKKOLA committed
June 21, 2026
A
cmake: drop ggml-blas.h from GGML_PUBLIC_HEADERS (#2007)
a1batross committed
June 19, 2026
K
Force Gemma4 assistant to be loaded on last GPU (#1999)
Kawrakow committed
K
Allow graph reuse for Gemma4 MTP (#1996)
Kawrakow committed
K
Fully remove any BLAS remnants (#2001)
Kawrakow committed
K
Add compatibility for llama.cpp Gemma4 assistant GGUFs (#1995)
Kawrakow committed
S
clean redudance in dflash graph and small logics (#1994)
Samuel Oliveira Alves committed
K
Fix Gemma4 MTP compute graph (#1993)
Kawrakow committed
K
Fix MTP warmup for GLM models (#1992)
Kawrakow committed
June 18, 2026
N
AVX VNNI auto-activation for MSVC ; HAVE_VNNI256 path for IQ4_XS_R8 and Qx_0 R4 quants. (#1991)
Nexes the Elder committed
K
Update AUTHORS
Kawrakow committed
F
faster ggml_cuda_host_malloc (#1988)
Farmadupe committed
K
Fix Qwen35 mtp warmup (#1987)
Kawrakow committed
June 17, 2026
K
Fix DFlash oerformance with split mode graph (#1980)
Kawrakow committed
June 16, 2026
J
Codex CLI Responses Compatibility (#1964)
Jun Yamog committed
J
chat: add Cohere2MoE North Code parser (#1968)
Joel Farthing committed
K
Merge pull request #1977 from ikawrakow/ik/dflash_fix_cpu
Kawrakow committed
K
Fix DFlash on the CPU
Kawrakow committed
K
Merge pull request #1970 from SamuelOliveirads/feat/dflash-implementation
Kawrakow committed
K
Merge pull request #1893 from ikawrakow/ik/gemma4_mtmd_blindness
Kawrakow committed
June 15, 2026
S
minor refactor in DFlash kv cache graph
SamuelOliveirads committed
K
Merge pull request #1973 from ikawrakow/ik/fattn_mma_gqa_16
Kawrakow committed
K
Merge pull request #1974 from Nexesenex/fix_muge_crash_minimax_m3
Kawrakow committed
N
Fix Minimax M3 crash when -muge merges up/gate experts
Nexesenex committed
K
CUDA FA: faster TG when GQA is 16 and head size is 128
Kawrakow committed
K
Merge pull request #1972 from ikawrakow/ik/minimaxm3_smgraph
Kawrakow committed
K
Merge pull request #1969 from Farmadupe/resize_algo_fix
Kawrakow committed