sam announced built-in support for decision models in Discourse AI, originally championed by TypeSafe.ai and now implemented by providers like CloudFlare and OpenAI. These models enable fast, accurate, and cost-effective classification of text and images by answering arbitrary questions. Unlike previous language model-based approaches, decision models offer superior speed and precision for classification tasks.
The feature is enabled via admin settings (visible only if Discourse Workflows is active) and exposes a new “Classify” node within workflows. Use cases include automatic moderation, team notifications, and auto-tagging. Users can configure their own classifiers to handle specific needs.
satonotdead highlighted the potential for response triage agents, asking if decision models could be integrated into common agents to create “Senior Helpers” that classify queries before delegating to specialized subagents. After sam expressed interest in exploring this as a routing gate, satonotdead detailed a proposed workflow where a decision model acts as a gate. This gate would select the appropriate subagent (e.g., publicator, web, base) based on existing subagent descriptions and user permissions. If confidence is high, the subagent runs directly; otherwise, the parent LLM reviews the classification as a hint. This approach could also serve as a guard to enforce safety policies (e.g., blocking trading signals in an education community) more reliably than prompt instructions alone.
Finally, sam noted that Discourse Cloud users have pre-configured access, while self-hosters can experiment by minting free keys at vllm-sr.ai.
I am very happy to announce we now have built-in support for decision models.
Decision models were championed and first developed by TypeSafe.ai and since then have been implemented by numerous providers including CloudFlare (clef), Open AI (decisions api) and many more.
They allow you to classify text and images by answering arbitrary questions about the inputs quickly, accurately and cheaply.
The new intelligence can be used for a very large amount of workflows given it allows you to classify text (and images) using your own designed classifier.
Automatic moderation - get notified when discussions get heated
Notify particular team members when particular topics are discussed
Automatic tagging and categorization
And much more
Enjoy, and please share interesting workflows you come up with.
I find this great for response triage agents; I’m already in my first tests to see how it goes. I understand that these new models for decision-making are very precise and fast, and I’m glad we can have them in Discourse.
I wonder if there are plans and/or if it is technically possible to incorporate them into common agents as well, to develop ‘Senior Helpers’ that first classify before delegating. Or does that not make sense at all?
It’s likely that I haven’t yet fully understood their real best use case, and maybe I’m asking something out of place.
I think there are a large number of uses here, using them as a gate that finds the right agent or subagent to answer is certainly something I want to explore
a chart a member shares, reviewed against our public method
Today the parent LLM reads the subagent descriptions and picks one with spawn_agent. The gate would take that decision from it:
A gate setting on the agent, pointing to a decision model.
The choices are the subagent descriptions that already exist, plus “answer yourself”. Nothing new for admins to write.
Only the subagents the current user can access are offered.
Above a confidence threshold, the subagent runs directly. Below it, the result reaches the parent LLM as a hint and the LLM decides, as it does now.
Since decision models classify images, “member shared a chart” comes from the image itself.
The same gate, as a yes/no question, would also work as a guard before the LLM runs.
We’re an education community and we don’t give signals, price targets or investment advice, so “is this asking for a trading signal?” → fixed reply. Today that rule is a line in every prompt, and models drift from it under pressure.
If you get to it, we’d be glad to try an early version on our forum.
