davsclaus opened a new issue, #124: URL: https://github.com/apache/camel-jbang-examples/issues/124
The ai group has a local model that writes text (langchain4j-chat), but no example where a model's answer drives the routing. A new example, `ai/ticket-classifier`, shows it with the web shop's support tickets: - a local model (granite4:3b on Ollama, from `camel infra` or installed) names the category of each ticket: shipping, billing or account, - the prompt is sent from a variable and the answer goes to one (`variableSend`, `variableReceive`), so the ticket stays the body, - the route does not trust the answer: it is normalized, and anything other than the three categories files the ticket under `unknown`, - a circuit breaker with a timeout files the ticket under `unknown` when the model is slow or down. It comes with a Citrus YAML test on the `camel infra` Ollama (`ciSkip`, like langchain4j-chat), and it is also a new task for the AI benchmark. _Claude Code on behalf of davsclaus_ -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
