⚡
Brief IA
›

LLM-Anthropic 0.30: Model Refresh and Token Counting

💻 Code & Dev·Tom Levy·

LLM-Anthropic 0.30: Model Refresh and Token Counting

LLM-Anthropic 0.30: Model Refresh and Token Counting
⚡
Key Takeaways
1llm-anthropic 0.30 allows updating the list of Anthropic models via the API
2A token counting command anticipates consumption before sending
3The publication date is September 28, 2026, at 11:06 PM
4Default budget caps are planned as the next step
💡Why it matters — These developments simplify model management and cost control for users.
⚡Le brief IA que lisent les pros

Le brief IA que les pros lisent chaque soir

Les 7 actus IA du jour, décryptées en 5 min. Gratuit.

Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.

Choisis ton rythme

Gratuit · Pas de spam · Désabonnement en 1 clic

Version 0.30 of the Anthropic integration focuses on client-side autonomy and request predictability. It introduces an automatic update of the model list and a token counting command, while setting a course towards default budget ceilings.

Release and Upcoming Budgetary Steps

Version 0.30 of llm-anthropic was released on September 28, 2026, at 11:06 PM. Among the upcoming steps, there are plans to establish default budget ceilings on nearly everything. This release comes in the context of current events marked by the announcement of a live blog for OpenAI DevDay 2026 on September 29, 2026, and a recap titled "2026 in LLMs (so far)" dated September 27, 2026.

Autonomous Update of the Model List

The command "llm anthropic refresh" allows for the updating of the Anthropic model list directly via their API. With this feature, it is no longer necessary to release a new version of the software to support a recently added model. This version also includes support for Claude Sonnet 5.5.

Anticipated Token Counting

A new command, "llm anthropic count," has been introduced. It uses Anthropic's free token counting API to indicate in advance the number of tokens a prompt will use, even before it is sent.

⚡

Brief IA — L'actualité IA en français

L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.