Release v0.3.0

• Add support for the following flags as optional arguments:
  • threads
  • top_k
  • top_p
  • repeat_last_n
  • repeat_penalty
  • ctx_size
  • ignore_eos
  • memory_f32
  • temp
  • n_parts
  • batch_size
  • keep
  • mlock

Release v0.2.1

• Update README.md

Release v0.2.0

• Compile and wrap llama.cpp's main instead of using the C++ code directly
• llama.cpp eeaa7b0492fc79baab8bb1fe195d6c87159f2bd3

Release v0.1.0

• Initial release
• llama.cpp 3525899277d2e2bdc8ec3f0e6e40c47251608700