0.9

• Added support for grounding prompts against Google search using the new -o google_search 1 option. Thanks, Ricardo Mestre. #29
• Added support for text/plain and text/csv attachments. Thanks, Michele Gregori. #34
• Fixed bug handling application/ogg attachments. #35
• New model: learnlm-1.5-pro-experimental. Thanks, Leonard Tulipan. #36
• New model: gemini-2.0-flash-thinking-exp-01-21. #37

0.8

• Support for Gemini 2.0 Flash Thinking Mode experimental model: llm -m gemini-2.0-flash-thinking-exp-1219 "Solve 3*x^3-5*x=1". #31
• Fixed bug with async models used in conversations. Thanks, Sukhbinder Singh. #30

0.7

• New Gemini 2.0 Flash model: llm -m gemini-2.0-flash-exp 'prompt goes here'. #28

0.6

• New model: gemini-exp-1206. #27

0.5

• Track token usage using new capability introduced in LLM 0.19. #25

0.4.2

• Support for new gemini-exp-1121 model. #26

0.5a0

• Track token usage. #25

0.4.1

• Depends on LLM 0.18 or higher. #23

0.4

• Handle attachments that are sent without a prompt. #20
• Support for new gemini-exp-1114 model. Thanks, Dominik Hayon. #21
• Support for JSON output mode using -o json_object 1. #22
• Now provides async versions of the Gemini models, compatible with LLM 0.18. #23

0.3

• Multi-modal model support with LLM 0.17 attachments. Gemini 1.5 models can now accept images, audio and video. #17

  llm -m gemini-1.5-flash-8b-latest 'describe image' \
    -a https://static.simonwillison.net/static/2024/pelicans.jpg

• Support for code execution mode. #18

  llm -m gemini-1.5-pro-latest 'write and execute python to calculate factorial of 13' -o code_execution 1

• Support for options: temperature, max_output_tokens, top_p, top_k. Pass these as e.g. -o temperature 0.5. #3