ngxson/llama.cpp

Pull Requests Commits

llama-fit-params: force disable mlock (#18103)

JohannesGaessler committed 50 minutes ago

Verified d0794e89

llama-fit-params: lower ctx size for multi GPU (#18101)

JohannesGaessler committed 51 minutes ago

Verified 9dcac6cf

llama-fit-params: fix underflow for dense models (#18095)

JohannesGaessler committed 52 minutes ago

Verified 0e49a7b8

llama-fit-params: QoL impr. for prints/errors (#18089)

JohannesGaessler committed 1 hour ago

Verified 4164596c

model: fix LFM2 missing tensors (#18105)

ngxson committed 6 hours ago

Verified ef83fb86

llama: fix early stop in params_fit if ctx is set (#18070)

JohannesGaessler committed 11 hours ago

Verified ec98e200

server: fix crash when batch > ubatch with embeddings (#17912)

yifant-code committed 12 hours ago

Verified 59977eba

model-conversion : remove -fa option in model card template [no ci] (#18088)

danbev committed 12 hours ago

Verified 79dbae03

arch: refactor LLM_TENSOR_NAMES (#18051)

ngxson committed 12 hours ago

Verified 7f2b2f3c

arg: clarify auto kvu/np being set on server (#17997)

ngxson committed 13 hours ago

Verified 7b1db3d3

Optimization: Qwen3 next autoregressive pass (#17996)

pwilkin committed 13 hours ago

Verified a5251ca1

CLI: fixed adding cli and completion into docker containers, improved docs (#18003)

andrew-aladev committed 13 hours ago

Verified fb644247

server: Update README.md incorrect argument (#18073)

2114L3 committed 13 hours ago

Verified 5f5f9b46

model: support GLM4V vision encoder (#18042)

ngxson committed 14 hours ago

Verified 3d86c6c2

model-conversion : add note about verifying previous models (#18082)

danbev committed 14 hours ago

Verified 9963b81f

model-conversion : use CONVERTED_EMBEDDING_MODEL for embedding_verify_logits (#18079)

danbev committed 14 hours ago

Verified db81d5ec

common : add nemotron 3 parsing (#18077)

aldehir committed 14 hours ago

Verified c05aa69f

added note for old Intel hardware pre sycl (#18017)

alosslessdev committed 14 hours ago

Verified 279cef27

security : add collaborator guidance (#18081)

ggerganov committed 15 hours ago

Verified 5ba95754

llama: Include algorithm header needed for C++23 (#18078)

cpeterso committed 17 hours ago

Verified 2aa45ef9

graph : reuse SSM graphs (#16490)

ggerganov committed 17 hours ago

Verified c5603164

ci : separate webui from server (#18072)

CISC committed 17 hours ago

Verified d6742125

webui: Improve copy to clipboard with text attachments (#17969)

allozaur committed 18 hours ago

Verified 3034836d

webui: Add setting to always show sidebar on Desktop (#17809)

allozaur committed 18 hours ago

Verified a20979d4

llama : add support for NVIDIA Nemotron 3 Nano (#18058)

danbev committed 18 hours ago

Verified 29953417

Webui: Disable attachment button and model selector button when prompt textbox is disabled. (#17925)

dariusjlukas committed 18 hours ago

Verified 40d9c394

convert : move rope_parameters to TextModel class (#18061)

CISC committed 1 day ago

Verified d6a1e18c

ggml-hexagon: mm for mtmd (#17894)

joeldushouyu committed 1 day ago

Verified c45f89d5

model : add KORMo model (#18032)

HelloKS committed 1 day ago

Verified 9d52f17a

kv-cache: Fix state restore fragmented cache (#17982)

ssweens committed 1 day ago

Verified 4529c660

Older