0.8.2

• Fix primary model name selection logic to prefer names with longer tags.
• Propagate input/output token usage information to llm.
  To see token usage, specify -u option, e.g.:

  $ llm -u -m llama3.2 "How much is 2+2?"
  The answer to 2 + 2 is 4.
  Token usage: 33 input, 13 output
  

0.8.1

• Fix bug in building messages in conversation when async models are used

0.8.0

• Add support for async LLM models

0.7.1

• Update plugin internals to be compatible with the latest 0.4.0 release of Ollama Python library

0.7.0

• Add support for text embedding.
  Example usage: llm embed -m mxbai-embed-large -i README.md
• Do not register embedding-only models (such mxbai-embed-large) for prompting and chatting

0.6.0

• Add support for image attachments.
  Example usage: llm -m llava "Describe this image" --attachment image.jpg

0.5.0

• Add support for forcing the model to reply with a valid JSON object

0.4.3

• Fix the type of stop option. This allows using it through the llm Python API; however, it's not clear how to pass it through the CLI.

0.4.2

• Ignore KeyError when iterating through response messages in streaming mode

0.4.1

• Prevent inability to communicate with Ollama server from failing the entire llm CLI