| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| llamacpp_for_kobold.zip | 2023-03-20 | 657.4 kB | |
| llamacpp-for-kobold-1.0.1 source code.tar.gz | 2023-03-20 | 2.0 MB | |
| llamacpp-for-kobold-1.0.1 source code.zip | 2023-03-20 | 2.0 MB | |
| README.md | 2023-03-20 | 400 Bytes | |
| Totals: 4 Items | 4.6 MB | 0 | |
llamacpp-for-kobold-1.0.1
- Bugfixes for OSX, and KV caching allows continuing a previous generation without reprocessing the whole prompt
- Weights not included.
To use, download, extract and run (defaults port is 5001):
llama_for_kobold.py [ggml_quant_model.bin] [port]
and then you can connect like this (or use the full koboldai client): https://lite.koboldai.net/?local=1&port=5001