llm 0.31

2026-04-24 · Source: Simon Willison's Weblog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

The `llm` command-line tool, version 0.31, was released on April 24, 2026, introducing several enhancements for interacting with OpenAI's large language models. Key updates include support for the new GPT-5.5 model, accessible via `llm -m gpt-5.5`. Users can now control text verbosity for GPT-5+ models using the `-o verbosity` option, with settings like `low`, `medium`, and `high`. Additionally, the release adds an `-o image_detail` option for image attachments, allowing `low`, `high`, `auto`, and `original` (for GPT-5.4 and 5.5) detail levels. The update also ensures that models specified in `extra-openai-models.yaml` are registered for asynchronous operations.

Key takeaway

For Machine Learning Engineers leveraging command-line tools for LLM interaction, this `llm` 0.31 update streamlines access to OpenAI's latest models and features. You should integrate the new GPT-5.5 model and experiment with the verbosity and image detail parameters to optimize your prompts and responses, especially when working with multimodal inputs or requiring concise outputs.

Key insights

The `llm` 0.31 release enhances command-line interaction with OpenAI models, adding GPT-5.5 support and new control options.

Method

The `llm` tool provides command-line access to LLMs, enabling model selection, text verbosity control, and image detail specification through dedicated options.

In practice

• Use `llm -m gpt-5.5` for the new OpenAI model.
• Set text verbosity with `-o verbosity low|medium|high`.
• Control image detail via `-o image_detail low|high|auto|original`.

Topics

• llm CLI
• GPT-5.5
• OpenAI Models
• Text Verbosity
• Image Detail Level

Code references

• simonw/llm

Best for: Machine Learning Engineer, AI Engineer, Software Engineer, NLP Engineer

Related on AIssential

• llm 0.32a2 — OpenAI's GPT-5 class models now use `/v1/responses` for interleaved reasoning, enhancing tool call visibility.
• llm 0.32a0 — The `llm` CLI tool provides direct command-line access to large language models.
• Sign of the future: GPT-5.5 — AI capabilities are rapidly accelerating across models, applications, and tool integration.
• The signal is the ceiling: Measurement limits of LLM-predicted experience ratings from open-ended survey text — Input text characteristics limit LLM prediction accuracy more than prompt or model choice.
• Prodigy v1.12: OpenAI integration, prompt engineering, task routers, deployment docs and more — Prodigy v1.12 integrates LLMs and advanced task routing to accelerate data annotation and prompt engineering workflows.
• IEEE Transactions on Artificial Intelligence, Volume 7, Issue 2, February 2026 — The volume showcases diverse AI advancements, from LLM applications in autonomous driving to specialized models for protein language processing.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Simon Willison's Weblog.