llm 0.32rc2 fixes a dependency issue, makes GPT-5.6 Luna the default model, and adds a new 'llm openai endpoint' command for running prompts against OpenAI-compatible endpoints without configuration.
# Release: llm 0.32rc2
Source: [https://simonwillison.net/2026/Jul/30/llm-rc2/](https://simonwillison.net/2026/Jul/30/llm-rc2/)
Hot on the heels of[RC1](https://simonwillison.net/2026/Jul/30/llm-rc1/), this fixes a dependency issue and also adds two neat new features:
> - The default model for users who have not set their own default is now[GPT\-5\.6 Luna](https://developers.openai.com/api/docs/models/gpt-5.6-luna)\. It was previously[GPT\-4o mini](https://developers.openai.com/api/docs/models/gpt-4o-mini)\. Luna is a much better and more recent model, albeit slightly more expensive \- $0\.20 per million input tokens and $1\.20 per million output tokens, compared to $0\.15/$0\.60 for 4o mini\. You can switch back to 4o mini using`llm models default gpt\-4o\-mini`, or switch to[GPT\-5 nano](https://developers.openai.com/api/docs/models/gpt-5-nano), an even cheaper default model \($0\.05/$0\.40\), using`llm models default gpt\-5\-nano`\.[\#1576](https://github.com/simonw/llm/issues/1576) - New[llm openai endpoint](https://llm.datasette.io/en/latest/other-models.html#openai-endpoint)command for running prompts, chats and model listings against arbitrary OpenAI\-compatible endpoints without first configuring a model\. These calls are not logged\.[\#1565](https://github.com/simonw/llm/issues/1565)
The`llm openai endpoint`command is*really*cool\. I got frustrated at the lack of an obvious CLI tool for trying out prompts against arbitrary OpenAI Chat Completions imitation endpoints, so I decided to add that to LLM itself\.
You don't even have to install LLM to use this\. Here's a`uvx`one\-liner for running a prompt \- with tools \- against an[LM Studio](https://lmstudio.ai/)local model:
```
uvx --pre llm openai endpoint http://127.0.0.1:1234/v1 \
T llm_version -T llm_time --td \
-m google/gemma-4-31b 'what is the current LLM version? And the time?'
```
[Output here](https://github.com/simonw/llm/pull/1568#issuecomment-5136163707)\.
The llm CLI tool version 0.32a2 has been released, adding support for OpenAI's /v1/responses endpoint to enable interleaved reasoning for GPT-5 class models.
The llm tool has been updated to version 0.32.1 to fix a dependency issue with the OpenAI library. A forthcoming 0.33 version will migrate from httpx to httpx2.
LLM 0.32rc1 introduces a redesigned message store with content-addressable hash IDs, enabling de-duplication and tree structures for forked conversations, plus support for new GPT-5.6 model variants.