TLDRocket
Sign in

llm-gemini 0.33

Simon Willison’s Weblog Simon Willison Covered by 7 sources

llm-gemini 0.33 adds support for Gemini 3.7 Flash and a few other new models. It also brings reasoning traces and server-side tools into LLM 0.32.

Based on reporting by Simon Willison’s Weblog, Simon Willison — read the original for the full story.

Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error

Simon Willison has pushed out llm-gemini 0.33, and this one is mainly about catching up with Google’s latest model releases. The plugin now supports Gemini 3.7 Flash, along with gemini-3.6-flash, gemini-3.5-flash-lite, and two embedding models: gemini-embedding-2 and gemini-embedding-001.

There’s more here than just a model list refresh. llm-gemini now works with LLM 0.32, which means users can see reasoning traces and turn on server-side tools. Willison even shows the pattern for invoking CodeExecution from the command line, with a Python example that calculates a factorial and multiplies it by 3.

He also tested Gemini 3.7 Flash by asking it to draw pelicans riding bicycles at different thinking levels. The high-effort result impressed him. Minimal, which had existed in 3.6 Flash, is gone in 3.7.

There was one odd snag, though. The pelican image he showed was rendered in Safari, and Firefox and Chrome displayed it differently because Safari is more forgiving of empty SVG filter elements. In those browsers the bicycle still appears, but the pelican disappears entirely. That’s the sort of tiny browser-specific nonsense that keeps generative image demos entertaining.

My take — AI-written commentary, not fact-checked reporting

This is the real shape of AI tooling now: not grand launches, just constant plumbing work so the newest model doesn’t sit there looking useless in a terminal. The boring upgrades matter more than the fireworks, which is irritating if one likes hype and very good if one likes software that actually runs.

Read more about this at: Simon Willison’s Weblog

Related stories

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.