latitude-dev/latitude-llm typescript-telemetry-4.2.0

TypeScript Telemetry v4.2.0

on GitHub

Added

• Bring your own LLM cost. capture(name, fn, { cost: { input, output } | { total } }) sets the
  cost (USD) of every LLM-call span inside the capture (never the capture's own wrapper span).
  new Latitude({ pricing: { "<provider>/<model>": { inputPer1M, outputPer1M } } }) prices LLM
  calls from their token counts, and new Latitude({ costResolver }) lets costResolver(usage)
  return a cost or undefined to fall back. setLlmCost(span, { input, output, total }) sets the
  cost of one span. Precedence: setLlmCost > capture cost > costResolver > pricing >
  Latitude's server-side pricing.
• Where the SDK sets cost it writes gen_ai.usage.input_cost, gen_ai.usage.output_cost and
  gen_ai.usage.total_cost plus latitude.cost.source="user", replacing any cost the
  instrumentation wrote on that span. total defaults to input + output, a total-only cost
  drops the instrumentation's input/output costs, and an explicit 0 is honoured. Cost is applied
  at export through an exporter wrapper, so other processors on a shared provider see the span
  unchanged. New exports: setLlmCost, ATTRIBUTES (now including costInput / costOutput /
  costTotal / costSource), COST_SOURCE_USER and the LlmCost, LlmUsage, ModelPricing
  and CostResolver types. LLM calls, provider, model and token counts are read from
  OpenTelemetry GenAI, OpenInference (including CrewAI's AGENT span), OpenLLMetry and Vercel AI
  SDK (ai.model.*, ai.usage.*) attributes, and cost is resolved when each span ends, before
  redaction.