llm-gemini 0.38.1 Simon Willison released llm-gemini 0.38.1, a dot release of the LLM plugin that fixes missing support for the `xhigh` reasoning level on Gemini 4 Hydrogen and Gemini 4 Neon; Gemini 4 Carbon supports only `low`, `medium` and `high`. Willison's updated comparison grid shows Gemini 4 Hydrogen at `high` used 5,311 output tokens and cost 3.99 cents, while the same model at `xhigh` used 31,477 output tokens, cost 23.61 cents, and produced a pelican competitive with Gemini 4 Neon. Release: llm-gemini 0.38.1 https://github.com/simonw/llm-gemini/releases/tag/0.38.1 Gemini 4 Carbon only supports low , medium and high reasoning levels - but Gemini 4 Hydrogen and Gemini 4 Neon both have an extra xhigh level. I missed that when I added those models to the llm-gemini plugin, which is now fixed in this dot release. The xhigh setting for Hydrogen makes a significant difference. I updated my Gemini 4 comparison grid https://static.simonwillison.net/static/2026/gemini-4-carbon-neon-hydrogen-with-xhigh.html - note how Hydrogen at high used 5,311 output tokens and cost 3.99 cents, but the same model at xhigh used 31,477 output tokens, cost 23.61 cents, and produced a pelican that's competitive with Gemini 4 Neon: Tags: gemini https://simonwillison.net/tags/gemini , llm https://simonwillison.net/tags/llm , llm-reasoning https://simonwillison.net/tags/llm-reasoning , pelican-riding-a-bicycle https://simonwillison.net/tags/pelican-riding-a-bicycle , ai https://simonwillison.net/tags/ai , llms https://simonwillison.net/tags/llms , google https://simonwillison.net/tags/google , generative-ai https://simonwillison.net/tags/generative-ai