Skip to content

Navigation Menu

Appearance settings

View all features
- BY COMPANY SIZE
  Enterprises
  Small and medium teams
  Startups
  Nonprofits
- BY USE CASE
  App Modernization
  DevSecOps
  DevOps
  CI/CD
  View all use cases
- BY INDUSTRY
  Healthcare
  Financial services
  Manufacturing
  Government
  View all industries
View all solutions
- EXPLORE BY TOPIC
  AI
  Software Development
  DevOps
  Security
  View all topics
- EXPLORE BY TYPE
  Customer stories
  Events & webinars
  Ebooks & reports
  Business insights
  GitHub Skills
- SUPPORT & SERVICES
  Documentation
  Customer support
  Community forum
  Trust center
  Partners
- COMMUNITY
  GitHub SponsorsFund open source developers
- PROGRAMS
  Security Lab
  Maintainer Community
  Accelerator
  Archive Program
- REPOSITORIES
  Topics
  Trending
  Collections
- ENTERPRISE SOLUTIONS
  Enterprise platformAI-powered developer platform
- AVAILABLE ADD-ONS
  GitHub Advanced SecurityEnterprise-grade security features
  Copilot for BusinessEnterprise-grade AI features
  Premium SupportEnterprise-grade 24/7 support
Pricing

Search code, repositories, users, issues, pull requests...

Search

Clear

Search syntax tips

Provide feedback

We read every piece of feedback, and take your input very seriously.

Include my email address so I can be contacted

Saved searches

Use saved searches to filter your results more quickly

Name

Query

To see all available qualifiers, see our documentation.

Appearance settings

You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session.

Dismiss alert

crc-org / llama.cpp Public

Notifications You must be signed in to change notification settings
Fork 1
Star 1

Code
Issues 1
Pull requests 2
Projects
Security
Insights

Additional navigation options

Code
Issues
Pull requests
Projects
Security
Insights

Work on the `graph_optimize` method#23

Open

kpouget wants to merge 56 commits intocrc-org:maincrc-org/llama.cpp:mainfrom

kpouget:rebase-b7356-optimizekpouget/crc-llama.cpp:rebase-b7356-optimizeCopy head branch name to clipboard

Conversation Commits56 (56)Checks Files changed

Open

Work on the `graph_optimize` method#23
kpouget wants to merge 56 commits intocrc-org:mainfrom
kpouget:rebase-b7356-optimize

Commits

Commits on Dec 9, 2025

CUDA: fix FP16 overflow in tile FA kernel (#17875)
JohannesGaessler
authored
CANN: add support for partial RoPE and Vision mode (#17543)
noemotiovon
authored
console: allow using arrow left/right, home/end keys and history mode (#17836)

ngxson
and
ServeurpersoCom
authored
model : nit, DeepSeek V1 MoE is 16B and GigaChat is 20B (#12652)
CISC
authored
cmake: fix Mach-O current version number (#17877)
Rhys-T
authored
ggml : allow fill node alloc inplace (#17870)
CISC
authored
metal : print node names for debugging (#17882)
ggerganov
authored
ggml : Provide macos-specific backtrace printing to avoid terminal death (#17869)
gabe-l-hart
authored
docs: clarify that CPU support should be first (#17886)
JohannesGaessler
authored
Add DIAG for CUDA (#17873)
pwilkin
authored
metal: SSM kernel improvements (#17876)

gabe-l-hart
and
ggerganov
authored
docs : update cpu and cuda ops (#17890)
CISC
authored
common : add parser for ministral/mistral large 3/devstral 2 (#17713)
aldehir
authored

Commits on Dec 10, 2025

fix softmax for iGPU (#17838)
NeoZhangJianyu
authored
convert: allow using quantized Mistral weight (#17889)

ngxson
and
compilade
authored
CUDA: fix unpadded strides in MMA FA kernel (#17891)
JohannesGaessler
authored
docs : update opencl ops (#17904)
lhez
authored
model : Qwen3-Next-80B-A3B has 48 layers (#17898)
EZForever
authored
cli: new CLI experience (#17824)

ngxson
and
bandoti
authored
cuda : add missing support check for xielu (#17895)
CISC
authored
ggml : remove GGML_KQ_MASK_PAD constant (#17910)
ggerganov
authored
Fix race conditions in threadpool when dealing with dynamic/frequent n_threads changes (#17748)
max-krasnyansky
authored
server: add presets (config) when using multiple models (#17859)

authored
cli: enable jinja by default (#17911)

ngxson
and
CISC
authored
mtmd: some small clean up (#17909)
ngxson
authored
ci: fix riscv64-native build (#17916)
CISC
authored
ggml-hexagon: fix `rope` failure at `test-backend-ops` (#17565)
chraac
authored

Commits on Dec 11, 2025

OWNERS: add file for OpenShift CI control
kpouget
committed
Add helper scripts
kpouget
committed
ggml: add the ggml-remotingfrontend and ggml-remotingbackend libraries
kpouget
committed
ggml: src: ggml-remotingfrontend/ggml-backend: add stub for .graph_optimize
kpouget
committed
src: llama-*: reduce the verbosity
kpouget
committed
tools: run: run: add timing instrumentation
kpouget
committed
ggml-metal: make less verbose
kpouget
committed
HACK: ggml-cpu: reduce the verbosity
kpouget
committed
ggml-remotingbackend: add missing includes
kpouget
committed
ggml/src/ggml-backend-reg.cpp: fix the frontend library name
kpouget
committed
ggml-remotingbackend/backend: disable the GGML_BACKEND_LIBRARY_METAL_DEVICE_CONTEXT lookup
kpouget
committed
ggml-remotingfrontend: disable USE_METAL_GUEST_SUPPORTS_OP
kpouget
committed
ggml-remotingfrontend/ggml-backend-reg: don't initialize the metal context when disabled
kpouget
committed
ggml-remotingfrontend: disable USE_FROM_PTR (for Linux)
kpouget
committed
ggml: src: ggml-remotingfrontend/ggml-remoting: use the arch to distinguish linux vs macos
kpouget
committed
Update to make it work with Linux
kpouget
committed
ggml: src: ggml-remotingfrontend/ggml-backend-buffer: don't stop on buffer_cpy_tensor(
kpouget
committed
ggml: force disable vulkan loader when compiling the ggml-remotingbackend
kpouget
committed
ggml-remoting: add support for the buffer_cpy_tensor function
kpouget
committed
ggml-remotingbackend: cleanups
kpouget
committed
ggml-remotingbackend: fix inconsist fatal message ...
kpouget
committed
ggml-remotingfrontend: properly distinguish if the backend supports buffer_from_host_ptr
kpouget
committed
Revert "metal : make the FA extra sizes consistent (#17143)"
kpouget
committed
ggml-remoting: add the code the the graph_optimize function
kpouget
committed
ggml-remoting: update the graph serial/deserial
kpouget
committed
Start integrating code for returning the optimized cgraph
kpouget
committed
Add code to track the frontend buffers
kpouget
committed
ggml-remotingfrontend: lookup the guest buffer from the host handle
kpouget
committed
HACK: ggml-backend-reg: allow disabling the Vulkan backend
kpouget
committed

Footer

© 2026 GitHub, Inc.

Footer navigation

Terms
Privacy
Security
Status
Community
Docs
Contact

You can’t perform that action at this time.