Distribute and run LLMs with a single file.
C++ cross-platformggufllama-cpplocal-ailocal-inferencelocal-llmopen-source-aisingle-file-executablespeech-to-text
Favor `-c 0` over `-c 6700` in README.md
Justine Tunneyb072b35eeaab
diff --git a/README.md b/README.mdindex 050bf6cd..ed835c00 100644--- a/README.md+++ b/README.md@@ -315,8 +315,8 @@ Here's an example of how you can use llamafile to summarize HTML URLs:echo '[/INST]') | llamafile \-m mistral-7b-instruct-v0.1.Q4_K_M.gguf \- -c 6700 \-f /dev/stdin \+ -c 0 \--temp 0 \-n 500 \--silent-prompt 2>/dev/null