ggerganov/llama.cpp

Pull Requests Commits

Add missing unrolls

pwilkin committed 1 hour ago

e78392f4

Revert accidental change.

pwilkin committed 1 hour ago

4b22b8ff

pwilkin committed 1 hour ago

a80cbd4f

Fix FATTN profiling

pwilkin committed 1 hour ago

ad26fb07

Converge implementation with export-graph-ops

pwilkin committed 1 hour ago

d1bf48dd

Add missing op parameters to the profiler; add support for test-backend-ops to run performance tests with exactly the tensor shapes from the run

pwilkin committed 1 hour ago

ac0665aa

docs, pass copy details

pwilkin committed 1 hour ago

438209f3

fix mul_mat_id stats, add throughput stat, add envvar trigger, add concurrent mode fix

pwilkin committed 1 hour ago

87af3770

fix builds, integrate vulkan profiler, fix copy events, fix export

pwilkin committed 1 hour ago

11df4991

Fix more missing backend stuff (and Python errors)

pwilkin committed 1 hour ago

f2468de1

add second dimension to reported tensors, fix Mac build, add missing initializer to all backends

pwilkin committed 1 hour ago

870accf7

feat: cool profiler thingy

pwilkin committed 1 hour ago

300d203d

common : remove hf cache migration (#23266)

angt committed 1 hour ago

Verified 45b455e6

ui: Update KaTeX package and clean up logs from `sass` warnings (#23275)

allozaur committed 2 hours ago

Verified 3a9c1b85

feat: add scroll-to-bottom button to chat + prevent forced scroll down (#23270)

allozaur committed 2 hours ago

Verified b9a2170f

ui: Refactor models store, MCP service, and gate logs behind VITE_DEBUG (#23236)

allozaur committed 2 hours ago

Verified 1ff0fc13

ui: Centralize monospace font styles in app.css (#23272)

allozaur committed 3 hours ago

Verified a135ec0b

webui: fix Tailwind v4 utility classes missing when built via cmake (#23253)

zoitrok committed 4 hours ago

Verified 232f4665

llama: initialize pre-norm embedding mask flag (#23256)

abetlen committed 5 hours ago

Verified 49c21f97

add myself to conversion (#23261)

CISC committed 6 hours ago

Verified 77e38d68

ci : added kleidiai-server to server-self-hosted workflow (#22435)

martin-klacer-arm committed 7 hours ago

Verified 053e01df

scripts : allow wc2wt with an existing branch (#23189)

ggerganov committed 10 hours ago

Verified c3f95c1f

sycl: scalar SWAR byte-subtract in Q6_K MMVQ dot product (#22156)

aicss-genai committed 11 hours ago

Verified 0caf2a1d

sycl: route small f32 matmuls to oneMKL, bypass oneDNN (#22150)

aicss-genai committed 11 hours ago

Verified 5511965b

sycl : fix error when use -mg 1 error (#23140)

arthw committed 11 hours ago

Verified e98bcfec

update bid to match each layers MTP source (#23237)

bit-incarnas committed 12 hours ago

Verified 1867a0c6

cmake : do not check for bin install dir (#23234)

CISC committed 16 hours ago

Verified dd7cad71

feat: Support d_conv=15 for ssm-conv.cu (#23017)

gabe-l-hart committed 19 hours ago

Verified 726704a1

cmake : fix LLAMA_BUILD_UI logic (#23190)

aldehir committed 22 hours ago

Verified 87589042

cmake : do not install conversion script (#23204)

CISC committed 1 day ago

Verified e0de4c24

Older