[
https://issues.apache.org/jira/browse/HDDS-15906?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Arafat Khan reassigned HDDS-15906:
----------------------------------
Assignee: Arafat Khan
> Add gateway LLM provider for Recon chatbot (OpenAI-compatible gateways)
> -----------------------------------------------------------------------
>
> Key: HDDS-15906
> URL: https://issues.apache.org/jira/browse/HDDS-15906
> Project: Apache Ozone
> Issue Type: Sub-task
> Components: Ozone Recon
> Reporter: Arafat Khan
> Assignee: Arafat Khan
> Priority: Major
> Fix For: 2.3.0
>
>
> *Problem*
> When customers route all LLM traffic through an OpenAI-compatible gateway
> (e.g. LiteLLM or similar), they must configure {{openai.*}} and put
> non-OpenAI model names under {{{}openai.models{}}}. That is confusing and
> error-prone.
> *Solution*
> Add a dedicated {{gateway}} provider that uses the same OpenAI-compatible
> client, with explicit config keys:
> * {{ozone.recon.chatbot.provider=gateway}}
> * {{ozone.recon.chatbot.gateway.api.key}}
> * {{ozone.recon.chatbot.gateway.base.url}} (required)
> * {{ozone.recon.chatbot.gateway.models}} (comma-separated aliases)
> Existing {{{}openai{}}}, {{{}gemini{}}}, and {{anthropic}} providers are
> unchanged.
> *Notes*
> * Model names in {{gateway.models}} must match the gateway’s configured
> aliases exactly.
> * No auto-discovery; admins maintain the model list in config.
> *Testing*
> * Unit tests for routing and dispatcher
> * Manual smoke test against an OpenAI-compatible gateway
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
