simonw/llm

Issues

• Implement an "install -U" option for Updating llm

  #670 opened a month ago by knbknb
  1

• `sqlite3.ProgrammingError: Error binding parameter 12: type 'ChatResponse' is not supported` when chatting with Ollama models

  #741 opened a month ago by heyajulia
  0

• please add deepseek-r1

  #719 opened 2 months ago by Kreijstal
  3

• Error: Client.__init__() got an unexpected keyword argument 'proxies'

  #709 opened 2 months ago by simonw
  8

• prompts in python are not saved it the database

  #738 opened a month ago by simonmaeldev
  0

• Plugins are gone after upgrade

  #733 opened a month ago by amitkot
  1

• Include file contents in `llm chat`

  #682 opened 3 months ago by jritsema
  4

• OpenAI o1 models require `max_completion_tokens` instead of `max_tokens`

  #724 opened a month ago by archer-eric
  3

• llm logs --prompts issue to see truncated prompts

  #736 opened a month ago by simonw
  2

• add --name TEXT option to 'llm prompt' command to set conversation_name

  #735 opened a month ago by rdslw
  0

• Having plugins installed slows even basic `llm --help` from 1.3s to 10.7s

  #732 opened a month ago by mcint
  1

• Disabling or hiding openai models while keeping openi compatible support

  #714 opened a month ago by oatmealm
  1

• [Feature Request] Option to Hide OpenAI Models

  #723 opened a month ago by jason-riddle
  1

• o3-mini: unsupported parameters listed in `docs/usage.md`

  #731 opened a month ago by bocytko
  0

• whisper-1 and llm whisper-api (fixes)

  #729 opened a month ago by leifulstrup
  0

• Support for o3-mini including reasoning_effort low/medium/high

  #728 opened a month ago by simonw
  5

• Feature request: ability to use MCP servers

  #696 opened 2 months ago by davidmoshal
  5

• No key found for openai when using local llamafile

  #726 opened a month ago by xorvoid
  0

• Add low cost affordable DeepInfra plugin

  #725 opened a month ago by codelabyz
  0

• Design annotations abstraction for responses that are not just a stream of plain text

  #716 opened 2 months ago by simonw
  17

• Dumping and reloading logs.db with sqlite-diffable FTS inconsistencies

  #680 opened 3 months ago by giuli007
  0

• [feature request] Add support for Chain of Thought (CoT) from deepseek-r1

  #722 opened a month ago by Rianico
  0

• `--xl/--extract-last` option

  #717 opened 2 months ago by simonw
  5

• -x/--extract option for returning content of first fenced code block

  #681 opened 3 months ago by simonw
  10

• Error while using async conversation

  #674 opened 2 months ago by sukhbinder
  4

• Much slower after upgrade to v0.20

  #715 opened 2 months ago by rothos
  1

• Provide documentation to set defaults

  #708 opened 2 months ago by igorgatis
  2

• Document model.attachment_types to show what attachment types a model can take

  #705 opened 2 months ago by nicoKoehler
  2

• `llm models` crashes if I close ollama

  #711 opened 2 months ago by ibehnam
  1

• Could not paste into chat multi line

  #713 opened 2 months ago by sluongng
  0

• Unofficial Plugin Directory

  #712 opened 2 months ago by luebken
  0

• Don't show default model when searching with `llm models -q`

  #710 opened 2 months ago by simonw
  1

• free copilot plugin

  #707 opened 2 months ago by bytearchive
  1

• How to collect prompt/response feedback?!

  #706 opened 2 months ago by luebken
  0

• Github models plugin

  #704 opened 2 months ago by tonybaloney
  0

• Amazon Bedrock Nova plugin

  #703 opened 2 months ago by kevinsimper
  1

• More uv documentation

  #702 opened 2 months ago by simonw
  1

• Support for new o1 model

  #676 opened 2 months ago by simonw
  13

• `llm models -q X` search option

  #700 opened 2 months ago by simonw
  2

• Support for Structured Output

  #675 opened 3 months ago by daniel-j-h
  2

• Feature request: global super system prompt for LLM (a profile)

  #694 opened 2 months ago by aborruso
  1

• Exception ignored in: <_io.TextIOWrapper name=4 mode='w' encoding='UTF-8'>

  #665 opened 3 months ago by purencool
  4

• Templates still require input argument, params alone not enough like docs suggest

  #684 opened 3 months ago by cmardiros
  0

• New gpt-4o-audio-preview models

  #677 opened 3 months ago by simonw
  4

• Add ability to unregister / not load default OpenAI plugin

  #671 opened 3 months ago by bcongdon
  1

• Use developer messages, not system messages, with OpenAI

  #678 opened 3 months ago by simonw
  0

• Warning: datetime.datetime.utcnow() is deprecated

  #672 opened 3 months ago by simonw
  1

• python API isn't seeing all models

  #668 opened 3 months ago by cleesmith
  1

• [FEATURE REQUEST] - integration with prompt eval tools

  #669 opened 3 months ago by kuatroka
  0

• llm.get_models() returns multiple instances of the same model

  #667 opened 3 months ago by simonw
  2