Jina AI (direct)

Direct Jina AI API calls (jina SDK or raw HTTP, no framework). Embeddings live at https://api.jina.ai/v1/embeddings and reranking at https://api.jina.ai/v1/rerank. Jina's embeddings endpoint is deliberately OpenAI-compatible — the vendor states it matches the input/output JSON schema of OpenAI's text-embedding-3-large model, so a customer already emitting gen_ai.* spans for an OpenAI-shaped call gets the same shape here for free. Jina has no admin/billing API (usage is dashboard-only under "API Key & Billing"), so spend is only capturable via this customer-side OTLP push. Token accounting is input-only: embeddings and rerank calls have no completion step, so there is no meaningful output_tokens metric — one that could only ever read zero would be worse than omitting it. input_tokens and total_tokens are what this connector reports instead. No telara-observability provider wrapper exists for Jina today (the package's optional-dependencies only cover langchain/langgraph/crewai/ autogen/llamaindex/openai-agents/openai/anthropic/azure-openai) — spans must come from the customer's own OTel instrumentation around the Jina SDK/HTTP call, or from the OpenAI instrumentation package if they call Jina through an OpenAI-compatible client. Spans should be tagged telara.span.kind=provider so the cost rollup excludes them when an active framework adapter already covers the same call (double-count guard).