LLM tool breaks on OpenAI SDK dependency shift
Simon Willison's LLM tool hit a breaking change when OpenAI's Python library removed its httpx dependency, exposing a fragile transitive dependency chain that broke fresh installations. The fix pins to an older OpenAI version, but the deeper shift matters: the upcoming 0.33 release will migrate to httpx2, signaling how tightly coupled developer tools remain to upstream library decisions. For practitioners building on OpenAI's SDK, this illustrates the ongoing brittleness of Python packaging in production AI workflows.
Modelwire context
ExplainerThe real issue isn't the 0.32.1 patch itself, but what it reveals: OpenAI's SDK maintainers removed a dependency without a major version bump, which is a packaging violation that cascaded into LLM users' environments. This wasn't a bug in LLM; it was a silent contract break upstream.
This story sits in the Python packaging reliability space, though we have no prior coverage to connect it to. The pattern it illustrates (tight coupling between developer tools and upstream SDK decisions) is endemic to the AI tooling layer right now. LLM's migration to httpx2 in 0.33 isn't a fix so much as a workaround that buys time until the next upstream shift breaks things differently. This belongs to the broader category of production fragility in AI workflows, not to any recent OpenAI announcement or capability shift.
If LLM 0.33 ships with httpx2 support before OpenAI releases a stable 2.0 version of their SDK that also targets httpx2, that confirms the pattern: LLM is chasing OpenAI's dependency choices rather than leading them. If fresh installs of LLM break again within six months due to another transitive dependency change, the underlying problem (lack of version pinning discipline upstream) remains unsolved.
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
MentionsSimon Willison · LLM · OpenAI · httpx · httpx2
Modelwire Editorial
This synthesis and analysis was prepared by the Modelwire editorial team. We use advanced language models to read, ground, and connect the day’s most significant AI developments, providing original strategic context that helps practitioners and leaders stay ahead of the frontier.
Modelwire summarizes, we don’t republish. Simon Willison originally reported this story as “llm 0.32.1”. The full content lives on simonwillison.net. If you’re a publisher and want a different summarization policy for your work, see our takedown page.