Post

Simon Willison
@simonwillison.net

New release of my llm-openai plugin supporting today's three new GPT-4.1 models from OpenAI: llm install -U llm-openai-plugin llm -m openai/gpt-4.1 "Generate an SVG of a pelican riding a bicycle" github.com/simonw/llm-o...

3 reposts55 likes

Simon Willison
@simonwillison.net

Not yet - the new OpenAI training cutoff date is May 2024, my first pelican benchmark was October 2024 And honestly, if they train on those images so far they'll continue to get some pretty awful pelicans!

Bornach
@bornach.bsky.social

"Astronaut riding a horse" was one of the earliest text-to-image challenges. You'd think they would have added every common creature riding some form of transport combo to the training data by now.

Simon Willison
@simonwillison.net

My notes on the GPT-4.1 series so far - 1m tokens (across all three models) and GPT-4.1-nano being their cheapest model yet are particularly notable simonwillison.net/2025/Apr/14/...

simonwillison.netGPT-4.1: Three new million token input models from OpenAI, including their cheapest model yetOpenAI introduced three new models this morning: GPT-4.1, GPT-4.1 mini and GPT-4.1 nano. These are API-only models right now, not available through the ChatGPT interface (though you can try them …
Simon Willison
@simonwillison.net

Added this note to my post to help illustrate how absurdly inexpensive these models have got now. I can use GPT-4.1 nano to generate descriptions of 4,000 images for less than a dollar!

Christian Köberl
@derkoe.dev

Thanks for the post! Just tried out 4.1 but like all other models that I've tried it struggles with figuring out the correct line numbers in diffs. For example if you want to review a unified diff, the line numbers often do not match the actual code. Only Claude gets it most of the time.