Pull requests: ggerganov/llama.cpp
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
SimpleChat: a simple and dumb web front end for testing /chat/completions and /completions end points and try chat
#7350
opened May 17, 2024 by
hanishkvc
Loading…
Fix floating point error with ndot progress printing in Perplexity and show stats with < 100 tasks
#7348
opened May 17, 2024 by
strawberrymelonpanda
Loading…
Another threadpool: Avoid creating hundreds of threads in GGML
performance
Speed related topics
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7342
opened May 17, 2024 by
besnardjb
Loading…
android : use "ci-android" branch for CI
android
Issues specific to Android
devops
improvements to build systems and github actions
review complexity : low
Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
#7341
opened May 17, 2024 by
ggerganov
Loading…
github-actions-labeler: initial commit
devops
improvements to build systems and github actions
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7330
opened May 16, 2024 by
mofosyne
Loading…
add Viking tokenizer support
model
Model specific
python
python script changes
review complexity : low
Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
#7329
opened May 16, 2024 by
jonabur
Loading…
Viking-7B tokenizer support
model
Model specific
python
python script changes
review complexity : low
Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
Fixed painfully slow single process builds.
build
Compilation issues
need feedback
Testing and feedback with results are needed
performance
Speed related topics
#7326
opened May 16, 2024 by
jboero
Loading…
Add support for larger Granite Code Models (20B, 34B)
model
Model specific
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7324
opened May 16, 2024 by
sroecker
Loading…
[SYCL] Update SYCL upscale operation
generation quality
Quality of model output
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
SYCL
https://en.wikipedia.org/wiki/SYCL - GPU programming language
#7321
opened May 16, 2024 by
AidanBeltonS
Loading…
sched : support async weight copy
performance
Speed related topics
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
ggml : fix quants nans when all the group weights are very close to zero
bugfix
fixes an issue or bug
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7313
opened May 15, 2024 by
slaren
Loading…
Add phi-2 tokenizer
model
Model specific
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7300
opened May 15, 2024 by
BramVanroy
Loading…
Capture CUDA logging output
enhancement
New feature or request
Nvidia GPU
Issues specific to Nvidia GPUs
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7298
opened May 15, 2024 by
fraxy-v
Loading…
avoid to get prompt in infill mode and embedding mode
examples
review complexity : low
Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
server
common: free ctx_gguf when exiting llama_control_vector_load_one
bugfix
fixes an issue or bug
review complexity : low
Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
#7285
opened May 14, 2024 by
stevegrubb
Loading…
ggml-opencl, llama: using reserve() if count already known
refactoring
Refactoring
review complexity : high
Generally require indepth knowledge of LLMs or GPUs
#7272
opened May 14, 2024 by
GermanAizek
•
Draft
common, ngram_cache: added const reference for std::pair<> and std::tuple<> more 16 bytes:
refactoring
Refactoring
review complexity : low
Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
#7270
opened May 14, 2024 by
GermanAizek
•
Draft
ggml, ngram-cache, log: added const and const ref for function params
refactoring
Refactoring
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7269
opened May 14, 2024 by
GermanAizek
Loading…
ggml llama: align structs for memory optimization on 64-bit platforms
refactoring
Refactoring
review complexity : medium
Generally require more time to grok but manageable by beginner to medium expertise level
#7267
opened May 13, 2024 by
GermanAizek
Loading…
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.