-
Notifications
You must be signed in to change notification settings - Fork 18.4k
Pull requests: ggml-org/llama.cpp
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Check the right iface method before using the fallback 2d get
ggml
changes relating to the ggml tensor library for machine learning
#23306
opened May 19, 2026 by
TheBlueMatt
Contributor
Loading…
fix: add Qwen3.5 MoE in_proj_qkv weight handling for SSM tensor export
python
python script changes
#23305
opened May 19, 2026 by
xmx-l
Loading…
opencl: add MoE support for q4_k, q5_k, q6_k on Adreno
ggml
changes relating to the ggml tensor library for machine learning
OpenCL
Issues specific to the OpenCL backend
#23303
opened May 18, 2026 by
shaofeiqi
Contributor
Loading…
ggml-webgpu: Fix new K>1 tests for GATED_DELTA_NET
ggml
changes relating to the ggml tensor library for machine learning
merge ready
A maintainer can use this label to indicate that they consider the changes final and ready to merge.
WebGPU
#23299
opened May 18, 2026 by
reeselevine
Contributor
Loading…
feat: add Vulkan REPEAT op support for f16 to f16.
ggml
changes relating to the ggml tensor library for machine learning
Vulkan
Issues specific to the Vulkan backend
#23298
opened May 18, 2026 by
l8bloom
Loading…
app : introduce the llama unified executable
build
Compilation issues
examples
server
#23296
opened May 18, 2026 by
angt
Member
Loading…
ui: silence a11y caption warning and tidy vitest setup
examples
server/ui
#23293
opened May 18, 2026 by
ServeurpersoCom
Contributor
Loading…
Move to backend sampling for MTP draft path
#23287
opened May 18, 2026 by
gaugarg-nv
Contributor
Loading…
common: fix --fit verbosity with --verbosity 4
examples
#23282
opened May 18, 2026 by
JohannesGaessler
Contributor
Loading…
server-context: fall back to full seq clear when partial KV eviction is refused
examples
server
#23280
opened May 18, 2026 by
ServeurpersoCom
Contributor
Loading…
server : print graphs reused in slot timings
examples
server
#23279
opened May 18, 2026 by
ggerganov
Member
Loading…
common: fix --help for --verbosity
#23278
opened May 18, 2026 by
JohannesGaessler
Contributor
Loading…
github: mention --log-file in issue templates
devops
improvements to build systems and github actions
#23277
opened May 18, 2026 by
JohannesGaessler
Contributor
Loading…
ui: prevent checkbox click from propagating in tools submenu
examples
server/ui
#23276
opened May 18, 2026 by
MaxKruse
Loading…
rpc : keep last_graph_uid in the device context
ggml
changes relating to the ggml tensor library for machine learning
merge ready
A maintainer can use this label to indicate that they consider the changes final and ready to merge.
#23273
opened May 18, 2026 by
rgerganov
Member
Loading…
ci : install server kleidiai runner dependencies
devops
improvements to build systems and github actions
#23259
opened May 18, 2026 by
CISC
Member
Loading…
Fix imatrix generation for MTP models
examples
python
python script changes
#23258
opened May 18, 2026 by
de-wim
Loading…
mtmd: add --mmproj-device argument
examples
server
#23255
opened May 18, 2026 by
Interpause
Loading…
4 tasks done
NvFp4 CT and Fp8 as Q8 conversion support
python
python script changes
#23250
opened May 18, 2026 by
ynankani
Contributor
Loading…
Add path validation and exit code error handling to bench script
script
Script related
#23248
opened May 18, 2026 by
Eamon2009
Loading…
common : support schema-constrained decoding for Gemma 4 tool calls
testing
Everything test related
#23247
opened May 18, 2026 by
rsauciuc
Loading…
rpc : track last graph uid per (endpoint, device), not per backend context
ggml
changes relating to the ggml tensor library for machine learning
#23243
opened May 18, 2026 by
ssam18
Contributor
Loading…
Previous Next
ProTip!
What’s not been updated in a month: updated:<2026-04-18.