LostRuins / koboldcpp

Run GGUF models easily with a KoboldAI UI. One File. Zero Install.

https://github.com/lostruins/koboldcpp

LostRuins/koboldcpp Issues

Program version number when using the -h command line argument
Updated 5 days ago
DRY sampling is broken in release version 1.73
Updated 5 days ago2
Repeated token issue
Closed 22 days ago12
[FEATURE Request] - DRuGS sampler: Deep Random Micro-Glitch Sampling
Updated 6 days ago1
Sponsorship Inquiry
Updated 6 days ago1
Out of VRAM even with zero GPU layers due to BLAS for llama-405 and mistral-large.
Updated 6 days ago1
Not able to run 32k context on v1.72.
Closed 6 days ago18
Pipeline parallelism?
Updated 6 days ago5
im on arm 64 bit rpi 5 whetres the pre compiled binary for thios for use with no gpu cpu only! fixit!
Updated 7 days ago15
Feature Request: Implement XTC sampler (new)
Updated 7 days ago1
On Linux, CUDA device names in the launch GUI are backwards from what they should be
Closed 7 days ago8
How to increase the generation limit in responses in apiopenai endpoint
Closed 7 days ago3
Cant install koboldcpp
Closed 7 days ago2
Stop sequence doesn't work
Closed 7 days ago2
This is not an issue but a feature request
Closed 7 days ago1
Request: Add input box to the idle duration in idle responses
Closed 7 days ago5
No longer possible to listen on all interfaces, which should be possible according to --help.
Updated 7 days ago2
Clicking stop will erase the progress
Updated 7 days ago4
SDUI negative_prompt URL parameter not working / implemented
Closed 9 days ago4
Quantkv CPU requirements?
Updated 9 days ago1
unknown model architecture: '' and SEGV while loading the model sdxl_lightning_4step.q8_0.gguf
Closed 11 days ago4
Horde worker not working under linux
Closed 11 days ago4
Missing SelfExtend options
Updated 12 days ago1
STATUS_BAD_FUNCTION_TABLE when running on nvidia gpu with vulkun
Updated 12 days ago1
SSE Stream add wait time input in GUI for the API
Updated 12 days ago3
After loading the model, click Launch and the following error appears
Closed 12 days ago8
Simple bug
Closed 13 days ago1
Problem with text generation
Closed 14 days ago2
C++ problems: use of uninitialized values, expressions are unused, comparisons always true, etc.
Updated 14 days ago1
Project doesn't install anything
Updated 14 days ago6
Getting a looping error during inference reading [control_76][control_76][control_76]...
Updated 15 days ago2
Multiple GPUs in Kobold PunPod?
Closed 18 days ago1
SSE streaming over API?
Closed 20 days ago2
How to use models?
Closed 20 days ago5
koboldcpp.exe crashes immediately after clicking the launch button. Help Me! I'm eager to start using koboldcpp
Closed 20 days ago1
Gpu acceleration on termux?
Updated 20 days ago2
row-split causing error
Updated 20 days ago
Instruct tag format accuracy
Updated 22 days ago4
How to hide <br><br>
Updated 22 days ago1
There is no sound
Updated 22 days ago11
Windows Vulkan backend approximately 50% slower with new Llama3 rope_factors
Updated 24 days ago4
Error using Kobold on Colab
Closed 25 days ago2
fatal error: 'execinfo.h' file not found
Closed a month ago2
L3.1 GGUFs made post Llama.cpp b3479 update not working in Kobold.cpp
Closed a month ago3
*SOLVED - CUDA fails after first run.
Closed a month ago1
OpenAI API endpoint has no way to enable "Assistant jailbreak"
Updated a month ago4
suggestion: provide one-folder pyinstaller for cuda builds
Closed a month ago5
Corpo theme cannot customise AI and user images
Updated a month ago
Unable to edit GUI text/ Can no longer see model selected
Closed a month ago1
Support ComfyUI as image generation backend
Updated a month ago1