--- title: New release of LLM adds support for reasoning traces,... description: LLM 0.32 is a major release of Simon Willison's LLM CLI and Python library. Key additions include visible reasoning traces sent to stderr, server-side tool... canonical: https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging | daily.dev og:description: LLM 0.32 is a major release of Simon Willison's LLM CLI and Python library. Key additions include visible reasoning traces sent to stderr, server-side tool... og:url: https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp og:image: https://api.daily.dev/og/posts/9YcxWoqVp.png og:image:alt: New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging **[Simon Willison](https://daily.dev/sources/simonwillison)** · 6 min read · 6 upvotes · 1 comments ## Summary LLM 0.32 is a major release of Simon Willison's LLM CLI and Python library. Key additions include visible reasoning traces sent to stderr, server-side tool support (OpenAI CodeInterpreter, WebSearch; Anthropic WebSearch, WebFetch, CodeExecution, AnthropicMCP), a new `llm openai endpoint` one-liner for hitting any OpenAI-compatible API, and a redesigned content-addressable SQLite logging system inspired by Git. The Python API gains a `messages=[]` parameter for passing full conversation history directly, plus a streaming events system that handles reasoning, text, tool calls, and image attachments as typed events. A new `llm-chat-completions-server` plugin exposes LLM as an OpenAI-compatible server. Updated plugins include llm-anthropic 0.26 with Claude 5 support. The release also positions LLM as an emerging agent framework, with tool chains that can pause for human approval and resume from stored history. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What new features does LLM 0.32 add for working with reasoning models and server-side tools? LLM 0.32 displays reasoning traces from reasoning models to standard error (disable with -R/--hide-reasoning), defaults to GPT-5.6 Luna, and supports server-side provider tools including OpenAI's CodeInterpreter and WebSearch, plus Anthropic's WebSearch, WebFetch, CodeExecution, and AnthropicMCP via the updated llm-anthropic 0.26 plugin. _Track fast-moving LLM tooling releases like this one on daily.dev before your agent stack falls behind._ ### How does LLM's new content-addressable message store work for logging conversations? It is a Git-modeled content-addressable message store that avoids re-logging duplicate JSON when a message sequence is appended to on every request, such as with chat completions style APIs. The llm logs and llm logs --json commands convert this format back into an easily consumable structure, documented in LLM's logging schema. _Developers wrestling with LLM conversation logging design can follow implementation details like this on daily.dev._ ### How do I run one-off prompts against a local OpenAI-compatible endpoint without installing LLM? Use uvx to run the llm openai endpoint command against any OpenAI-compatible API without a full LLM installation, for example targeting a local LM Studio server running Gemma 4 12B. These endpoint calls are not logged, making the command useful for quick one-off prompts, and tool plugins like llm-tools-quickjs can be mixed in via --with. _Anyone testing local model endpoints can keep up with CLI tooling changes like this via daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@agustinbarrientos** · 0 upvotes > Do you see the agent loop becoming a first-class abstraction in LLM, or do you want the event protocol to stay flexible enough for people to build their own? ## Similar posts on daily.dev - [LLM 0.32a0 is a major backwards-compatible refactor](https://daily.dev/posts/llm-0-32a0-is-a-major-backwards-compatible-refactor-5wwlmcnzs) · Simon Willison · 1 upvotes · 0 comments --- Tags: [#python](https://daily.dev/tags/python), [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#openai](https://daily.dev/tags/openai), [#mcp](https://daily.dev/tags/mcp) [View this post on daily.dev](https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging","url":"https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp"},"datePublished":"2026-08-05T00:03:33.757Z","dateModified":"2026-09-13T20:33:01.633Z","description":"LLM 0.32 is a major release of Simon Willison's LLM CLI and Python library. Key additions include visible reasoning traces sent to stderr, server-side tool...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4ea2f7b102469ce7997afaeb8d60510f?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4ea2f7b102469ce7997afaeb8d60510f?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Simon Willison","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Simon Willison","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/cf12b897300740218b35f49a6309ec5b","url":"https://daily.dev/sources/simonwillison"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":6},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"python,llm,ai-agents,openai,mcp","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Simon Willison","item":"https://daily.dev/sources/simonwillison"},{"@type":"ListItem","position":3,"name":"New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp","comment":[{"@type":"Comment","text":"Do you see the agent loop becoming a first-class abstraction in LLM, or do you want the event protocol to stay flexible enough for people to build their own?","datePublished":"2026-08-05T20:53:01.721Z","url":"https://daily.dev/posts/9YcxWoqVp#c-TO3txUJrg","author":{"@type":"Person","name":"Agustin Barrientos","url":"https://daily.dev/agustinbarrientos","image":"https://media.daily.dev/image/upload/s--5ayxQnqn--/f_auto/v1788281802/avatars/avatar_wQYYVe5Tbj0NJ7C7qPoa8?_a=BAMAMicg0"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/new-release-of-llm-adds-support-for-reasoning-traces-openai-responses-server-side-tools-and-smart-9ycxwoqvp#faq","mainEntity":[{"@type":"Question","name":"What new features does LLM 0.32 add for working with reasoning models and server-side tools?","acceptedAnswer":{"@type":"Answer","text":"LLM 0.32 displays reasoning traces from reasoning models to standard error (disable with -R/--hide-reasoning), defaults to GPT-5.6 Luna, and supports server-side provider tools including OpenAI's CodeInterpreter and WebSearch, plus Anthropic's WebSearch, WebFetch, CodeExecution, and AnthropicMCP via the updated llm-anthropic 0.26 plugin. Track fast-moving LLM tooling releases like this one on daily.dev before your agent stack falls behind."}},{"@type":"Question","name":"How does LLM's new content-addressable message store work for logging conversations?","acceptedAnswer":{"@type":"Answer","text":"It is a Git-modeled content-addressable message store that avoids re-logging duplicate JSON when a message sequence is appended to on every request, such as with chat completions style APIs. The llm logs and llm logs --json commands convert this format back into an easily consumable structure, documented in LLM's logging schema. Developers wrestling with LLM conversation logging design can follow implementation details like this on daily.dev."}},{"@type":"Question","name":"How do I run one-off prompts against a local OpenAI-compatible endpoint without installing LLM?","acceptedAnswer":{"@type":"Answer","text":"Use uvx to run the llm openai endpoint command against any OpenAI-compatible API without a full LLM installation, for example targeting a local LM Studio server running Gemma 4 12B. These endpoint calls are not logged, making the command useful for quick one-off prompts, and tool plugins like llm-tools-quickjs can be mixed in via --with. Anyone testing local model endpoints can keep up with CLI tooling changes like this via daily.dev."}}]} ```