AI tracing: Other providers
Any library that writes OpenTelemetry GenAI attributes works. OpenLLMetry covers many providers at once.
1. OpenLLMetry: one package for many providers
It instruments Mistral, Ollama, Cohere, Groq, Together, Bedrock, Vertex AI and more, and sends to PowerTraces directly.
pip install traceloop-sdk
from traceloop.sdk import Traceloop
Traceloop.init(
app_name="support-assistant",
api_endpoint="https://<address>.powertraces.com",
headers={"X-Otel-Api-Key": "<server key>"},
)2. Your own code: set the GenAI attributes
Wrapping a provider yourself? Put these attributes on the span around each model call; agent and tool spans use gen_ai.agent.name and gen_ai.tool.name.
with tracer.start_as_current_span("chat mistral-large") as span:
span.set_attribute("gen_ai.operation.name", "chat")
span.set_attribute("gen_ai.provider.name", "mistral_ai")
span.set_attribute("gen_ai.request.model", "mistral-large")
reply = client.chat(...)
span.set_attribute("gen_ai.usage.input_tokens", reply.usage.prompt_tokens)
span.set_attribute("gen_ai.usage.output_tokens", reply.usage.completion_tokens)Using other attribute names? Map them in Attribute mappings › AI, and set prices for models without a built-in price in AI › Settings.
Prompts and answers
Most libraries send only counts and timings until you turn content on. When you do, PowerTraces keeps it, masks personal data, or drops it, as set in AI › Settings for each workspace.
Check it works
Make one model call, then open the AI page: the run appears within a minute. Open it to see each step in the Agent tab.