higress · Issues· 1132 open

Proposal: support configuration-defined prompts in mcp-server

AI 模型路由插件model-router里面的自动匹配模型多模态不支持的问题 || The automatic matching model multi-modality in the AI model routing plug-in model-router does not support the problem

Design: Responses content extraction in ai-statistics (#3581)

Design: Integer/Short Dubbo string parameter conversion (#3659)

Proposal: make multi-model fallback progress through backup models (#3930)

Proposal: support Integer and Short Dubbo string parameters (#3659)

Proposal: collect Responses API answer and summary text (#3581)

Proposal: ext-auth: cache auth results, decide allow/deny from response content, and forward response fields upstream

Design: Preserve API identity across ai-proxy retry transforms (#4396)

Proposal: Preserve resolved API identity in ai-proxy retries (#4396)

Design: ext-auth: cache auth results, decide allow/deny from response content, and forward response fields upstream

[ai-proxy] qwen provider 多模态请求仅支持 qwen-vl 前缀模型，其他多模态模型报 InvalidParameter || [ai-proxy] qwen provider multi-modal request only supports qwen-vl prefix model, other multi-modal models report InvalidParameter

Design: isolate model-router auto-routing text to the final user message

Design: configurable model-router auto-routing trigger model ID

Proposal: prevent model-router auto routing from reusing stale text across user turns