issues
search
LostRuins
/
koboldcpp
A simple one-file way to run various GGML and GGUF models with KoboldAI's UI
https://github.com/lostruins/koboldcpp
GNU Affero General Public License v3.0
4.32k
stars
310
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Experimental
#958
Fischherboot
closed
25 minutes ago
0
Resolve `make: *** No rule to make target `ggml-metal.m', needed by `…
#957
beebopkim
closed
13 hours ago
1
Chat Adapters
#956
henk717
closed
19 hours ago
1
Feature Request: Support for Intel® Arc™ Graphics
#955
atlatlir
opened
1 day ago
1
Remove some old Cuda implementation from KCPP and realign on LCPP
#954
Nexesenex
closed
1 day ago
2
large models randomly don't work???
#952
pl752
closed
1 day ago
3
IDK chat Bug? kobold give random code...
#951
alastor367
closed
2 days ago
3
Gemma-2 support?
#950
pbz134
opened
2 days ago
2
Show this while loading
#949
lolo17787
opened
3 days ago
3
CUDA: fix MMQ stream-k for --split-mode row #8167
#948
Nexesenex
closed
19 hours ago
1
Add Cuda Graphs in CMakeList
#947
Nexesenex
closed
1 day ago
2
Add Cublas12 dlls to .gitignore
#946
Nexesenex
closed
2 days ago
0
CMakeList - Remove deprecated MMQ_Y tile size param
#945
Nexesenex
closed
4 days ago
0
Delete CMakePresets.json
#944
Nexesenex
closed
5 days ago
0
Feature request: DRY support
#943
AphidGit
opened
5 days ago
1
Add CMake flag for pipeline parallelism for multi-GPU
#940
Nexesenex
closed
5 days ago
0
Add OpenMP support in CMakeList
#939
Nexesenex
closed
5 days ago
0
Gradient rope formula with offsets
#938
Nexesenex
closed
5 days ago
0
Give the CI builds a recognizable AVX1 name
#937
henk717
closed
5 days ago
0
CUDA error when trying to run with hipblas
#936
ockerman0
opened
1 week ago
2
Unclear error message when provide wrong gbnf
#935
Kas1o
opened
1 week ago
0
Something breaks and then only gibberish is generated for DeepSeek-Coder-V2-Lite-Instruct
#933
aleksusklim
opened
1 week ago
1
Phi-3 vulkan?
#931
paryska99
opened
1 week ago
0
Feature request: minimum context tokens
#930
Azirine
opened
1 week ago
4
Augmented benchmark stats
#929
Nexesenex
closed
1 week ago
2
Is there a way to share story as plaintext?
#928
AlexysLovesLexxie
opened
1 week ago
4
Embed manifest.json from lite
#926
jojorne
closed
2 weeks ago
9
Auto apply correct chat template when using OpenAI compatible API
#925
Tureti
closed
2 weeks ago
3
Help: How to set Phi-3 prompt template?
#924
LeMoussel
closed
2 weeks ago
1
Open-web-ui not working with Kobold
#922
Tureti
closed
2 weeks ago
1
Cannot initialize microphone (non-localhost URL)
#920
CorentinWicht
opened
2 weeks ago
8
SDUI width/height parameters having no effect
#919
CorentinWicht
closed
2 weeks ago
8
Check for keyanti before accessing it
#918
valadaptive
closed
2 weeks ago
1
Feature request: Option to disable auto adding BOS token (double BOS token) if it's already present/added.
#917
Spacellary
opened
2 weeks ago
6
Option to selectively disable GPU acceleration for each model type
#916
tororon1231
closed
2 weeks ago
4
Build failed on macOS with concedo_experimental branch
#915
beebopkim
closed
2 weeks ago
5
Reduced generation speed in 1.67
#913
Vladonai
opened
2 weeks ago
1
bark.cpp as TTS
#912
tororon1231
closed
2 weeks ago
7
Maybe add gguf fine-tuning function and can be used in ui.
#911
win10ogod
opened
2 weeks ago
0
GradientAI Auto ROPE Base calculation
#910
askmyteapot
closed
2 weeks ago
4
Qwen2-72B-Instruct generates random output
#909
EugeoSynthesisThirtyTwo
opened
2 weeks ago
16
New requirements: 1. I hope to be able to load the webpage first like the open webui, and the model can be freely selected and switched in the webui. 2. When not used after 5 minutes or any set time, all GPU memory should be automatically released and loaded the next time someone requests an inference task on the web.
#907
windkwbs
closed
2 weeks ago
1
--quantkv error with Metal
#906
Azirine
opened
3 weeks ago
1
Better Undo/Redo
#905
Azirine
closed
1 week ago
5
Crash when loading GPU such as q_XS and q_XSS models (Preset NoAVX2)
#904
grtsata
opened
3 weeks ago
7
Feature request: Ability to "Generate More" when using Speech Control
#903
NGgmnBQGwR
closed
1 week ago
4
"Detect Voice" mode doesn't work in Firefox
#902
NGgmnBQGwR
closed
1 week ago
3
some non-ascii character streaming issue
#901
luzamm
opened
3 weeks ago
1
Flash attention slower
#900
Azirine
closed
3 weeks ago
4
Creating your own instruct Tag Preset
#899
berkut1
opened
3 weeks ago
1
Next