diff --git a/koboldcpp.py b/koboldcpp.py index 0e11d72a6..6f602b9f9 100755 --- a/koboldcpp.py +++ b/koboldcpp.py @@ -8261,7 +8261,7 @@ if __name__ == '__main__': advparser.add_argument("--ignoremissing", help="Ignores all missing non-essential files, just skipping them instead.", action='store_true') advparser.add_argument("--chatcompletionsadapter", metavar=('[filename]'), help="Select an optional ChatCompletions Adapter JSON file to force custom instruct tags.", default="AutoGuess") advparser.add_argument("--jinja", help="Enables using jinja chat template formatting for chat completions endpoint. Other endpoints are unaffected. Tool calls are done without jinja.", action='store_true') - advparser.add_argument("--jinja_tools","--jinja-tools", help="Enables using jinja chat template formatting for chat completions endpoint. Other endpoints are unaffected. Tool calls are done with jinja.", action='store_true') + advparser.add_argument("--jinja_tools","--jinja-tools","--jinjatools", help="Enables using jinja chat template formatting for chat completions endpoint. Other endpoints are unaffected. Tool calls are done with jinja.", action='store_true') advparser.add_argument("--flashattention","--flash-attn","-fa", help="Enables flash attention.", action='store_true') advparser.add_argument("--lowvram","-nkvo","--no-kv-offload", help="If supported by the backend, do not offload KV to GPU (lowvram mode). Not recommended, will be slow.", action='store_true') advparser.add_argument("--quantkv", help="Sets the KV cache data type quantization, 0=f16, 1=q8, 2=q4. Requires Flash Attention for full effect, otherwise only K cache is quantized.",metavar=('[quantization level 0/1/2]'), type=int, choices=[0,1,2], default=0)