Email or username:

Password:

Forgot your password?
Simon's posts Post Back to profile
Simon Willison

New release of my LLM combined CLI tool and Python library for interacting with LLMs - the big new feature in 0.18 is support for async models llm.datasette.io/en/stable/cha

0.18 (2024-11-17) Initial support for async models. Plugins can now provide an AsyncModel subclass that can be accessed in the Python API using the new llm.get_async_model(model_id) method. See async models in the Python API docs and implementing async models in plugins. #507 OpenAI models all now include async models, so function calls such as llm.get_async_model("gpt-4o-mini") will return an async model. gpt-4o-audio-preview model can be used to send audio attachments to the GPT-4o audio model. #608 Attachments can now be sent without requiring a prompt. #611 llm models --options now includes information on whether a model supports attachments. #612 llm models --async shows available async models. Custom OpenAI-compatible models can now be marked as can_stream: false in the YAML if they do not support streaming. Thanks, Chris Mungall. #600 Fixed bug where OpenAI usage data was incorrectly serialized to JSON. #614 Standardized on audio/wav MIME type for audio attachments rather than audio/wave. #603
17 November at 20:48 | Open on fedi.simonwillison.net
6 comments
Simon Willison

And a new plugin release: llm-claude-3 0.9, adding support for asynchronous access to the Claude family of models github.com/simonw/llm-claude-3

17 November at 21:06 | Open on fedi.simonwillison.net
Simon Willison

And here's llm-gemini 0.4, adding asynchronous support for the Google Gemini models plus the new Chatbot Arena-topping gemini-exp-1114 (and -o json_object 1 JSON mode for good measure) github.com/simonw/llm-gemini/r

0.4 Latest @simonw simonw released this 4 minutes ago 0.4 2c2eb63 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
18 November at 7:40 | Open on fedi.simonwillison.net
Simon Willison

llm-mistral 0.8 adds async model support for Mistral models, including the new Pixtral Large github.com/simonw/llm-mistral/

0.8 Latest @simonw simonw released this 19 Nov 00:56 0.8 f590da3 Provide async versions of the Mistral models, compatible with LLM 0.18. #13 Added a pixtral-large alias for the new Pixtral Large model. #14 llm -m pixtral-large describe -a https://static.simonwillison.net/static/2024/pelicans.jpg
19 November at 0:57 | Open on fedi.simonwillison.net
sayrer

@simon do you know what “Mistral" means? I think it is so funny. No judging, it could be “of course" or “what?”

19 November at 1:13 | Open on mastodon.social
Simon Willison

@sayrer does it mean something other than a strong cold dry northerly wind?

19 November at 4:38 | Open on fedi.simonwillison.net
sayrer

@simon yes, it's this one: en.wikipedia.org/wiki/Mistral_

But it is persistent (unrelenting?). That's why it is not just vent/wind. Like, it will still take your hat off tomorrow. Not a French word, really. It's something I would hear old guys complaining about in Provençal when I was very young there.

In the context of an AI company, I take it to mean “progress will happen”

19 November at 5:03 | Open on mastodon.social
Go Up