OpenAI Decisions API gains community plugin support with image capabilities
Source published ·Modelwire updated
Original coverage: Simon Willison ↗·How Modelwire adds context
The development
OpenAI's new Decisions API, unveiled at DevDay, enables structured decision-making workflows comparable to Anthropic's Jev system. Simon Willison rapidly built an llm-openai-decisions plugin that wraps the API for the LLM CLI ecosystem, leveraging GPT-6 Astra to reverse-engineer the documentation. The gpt-6-luna decision model adds multimodal input support absent in competing systems, signaling OpenAI's push to embed reasoning and choice logic into production workflows. This matters for developers integrating deterministic decision trees into agentic systems.
Modelwire’s AI-generated summary of coverage from Simon Willison.
Modelwire analysis
Analyst takeOur AI-generated reading of the wider context and the next developments to watch.
Willison's plugin shipped within hours of the API announcement, suggesting the Decisions API has lower friction than competing systems. The fact that gpt-6-luna's multimodal support is being highlighted as differentiation implies other decision models (Anthropic's Jev, AWS's Strands Decider) may lack this capability, making input format a new axis of vendor differentiation.
This release accelerates the pattern established in early October: Amazon's Strands Decider and Musubi's PolicyLM both signaled that decision models are becoming specialized, task-specific infrastructure rather than general-purpose tools. OpenAI's strategy here mirrors that fragmentation but with a twist. Where AWS and others are building proprietary models, OpenAI is embedding decision logic into its frontier model family (GPT-6 Luna) and betting that rapid plugin adoption by developers like Willison will create switching costs. The llm-mistral plugin release the same day shows the open tooling ecosystem is moving in parallel, but Willison's immediate OpenAI plugin suggests OpenAI's API surface is winning the integration race.
If Willison's llm-openai-decisions plugin reaches 1000+ GitHub stars or gets vendored into the core llm CLI within 30 days, that signals OpenAI's API design is sticky enough to outpace Anthropic's Jev adoption rate. If competing decision models (Strands, PolicyLM) don't ship equivalent multimodal support within Q4 2026, OpenAI will have established a defensible product moat beyond just model quality.
This interpretation is generated from the summary above and the archive coverage cited below. Our methodology · Report an error
Coverage behind this analysis
These archive entries ground the connection in our analysis. They are ordered by source publication date, with links to our coverage and the original sources.
·TechCrunch - AI
AWS launches decision model as enterprise reasoning tools splinter
AWS's Strand Labs has entered the competitive decision-model space with Strands Decider 2B, joining a wave of enterprise players building specialized reasoning systems. The release signals that major cloud providers are no longer content licensing third-party models; they're building proprietary inference layers tailored to operational workflows. This fragmentation of the decision-model category mirrors earlier consolidation…
MentionsOpenAI · Simon Willison · GPT-6 Astra · gpt-6-luna · llm-openai-decisions · Decisions API
How this coverage is produced
Modelwire uses AI to generate summaries and context from source headlines, snippets, and selected archive coverage. Automated checks do not verify every claim, and items are not routinely reviewed by a person before publication. Zacaria Solis operates the site. Read the linked source for the full evidence and report errors through our corrections process.
Modelwire summarizes, we don’t republish. Simon Willison originally reported this story as “llm-openai-decisions 0.1a0”. 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.