LLM observability

LLM observability turns a Vercel AI SDK call inside a task into its own span in the run trace, next to your logs and other spans. Each span carries the model, provider, input, output, and total token counts, cost, and latency, so you can see what each generation did and what it cost without leaving the run.

Everything shows up inline in the run trace you already use to debug runs. There is no separate product and no dashboard to set up.

Turn it on

Set experimental_telemetry: { isEnabled: true } on the AI SDK call. There is nothing to install for AI SDK 6, and nothing to configure on the Trigger.dev side.

import { task } from "@trigger.dev/sdk";
import { generateText } from "ai";
import { openai } from "@ai-sdk/openai";

export const summarize = task({
  id: "summarize",
  run: async (payload: { text: string }) => {
    const result = await generateText({
      model: openai("gpt-4o"),
      prompt: `Summarize the following text:\n\n${payload.text}`,
      experimental_telemetry: { isEnabled: true },
    });

    return { summary: result.text };
  },
});

Trigger the task and open the run. The generateText call appears as a span in the trace. streamText and generateObject work the same way: add the same experimental_telemetry flag to each call you want captured.

import { registerTelemetry } from "ai";
import { OpenTelemetry } from "@ai-sdk/otel";

registerTelemetry(new OpenTelemetry());

A chat.agent() run registers the adapter for you at run start, so chat agents need only the install. On AI SDK 5 and 6, ai core emits spans directly and no adapter is needed. </Note>

What each span shows

Open an AI generation span in the run trace to get a dedicated inspector with three tabs:

• Overview: model, provider, token usage, cost, and a preview of the input and output.
• Messages: the full message thread, including the system prompt and any tool results.
• Tools: the tool definitions passed to the model, plus every tool call the model made with its arguments.

A fourth Prompt tab appears when the call is linked to an AI Prompt (see below).

Link a call to its prompt

If you manage prompts with AI Prompts, resolve the prompt and spread toAISDKTelemetry() into the call. This sets experimental_telemetry for you and links the span back to the exact prompt version that produced it.

import { task, prompts } from "@trigger.dev/sdk";
import { generateText } from "ai";
import { openai } from "@ai-sdk/openai";
import type { supportPrompt } from "./prompts";

export const handleSupport = task({
  id: "handle-support",
  run: async (payload: { name: string; plan: string; issue: string }) => {
    const resolved = await prompts.resolve<typeof supportPrompt>("customer-support", {
      customerName: payload.name,
      plan: payload.plan,
      issue: payload.issue,
    });

    const result = await generateText({
      model: openai(resolved.model ?? "gpt-4o"),
      system: resolved.text,
      prompt: payload.issue,
      ...resolved.toAISDKTelemetry(),
    });

    return { response: result.text };
  },
});

The span's Prompt tab now shows the linked template, its version, and the input variables the prompt was resolved with.

Pass custom attributes to toAISDKTelemetry() to tag the span with your own metadata:

const result = await generateText({
  model: openai(resolved.model ?? "gpt-4o"),
  system: resolved.text,
  prompt: payload.issue,
  ...resolved.toAISDKTelemetry({
    "task.type": "summarization",
    "customer.tier": "enterprise",
  }),
});

Custom attributes are stored on the span's metadata, so you can filter or group by them in TRQL, for example metadata['task.type'].

Query usage across runs

Every captured generation is also written to the llm_metrics table, which you can query with TRQL. This lets you aggregate token usage, cost, and latency across many runs rather than inspecting one span at a time.

Cost and token usage by model:

SELECT
  response_model,
  gen_ai_system AS provider,
  count() AS calls,
  sum(total_tokens) AS tokens,
  round(sum(total_cost), 4) AS cost_usd
FROM llm_metrics
GROUP BY response_model, gen_ai_system
ORDER BY cost_usd DESC
LIMIT 20

Spend per task:

SELECT
  task_identifier,
  sum(input_tokens) AS input_tokens,
  sum(output_tokens) AS output_tokens,
  round(sum(total_cost), 4) AS cost_usd
FROM llm_metrics
GROUP BY task_identifier
ORDER BY cost_usd DESC
LIMIT 20

Cost by prompt version, when calls are linked to an AI Prompt:

SELECT
  prompt_slug,
  prompt_version,
  count() AS calls,
  round(sum(total_cost), 4) AS cost_usd
FROM llm_metrics
WHERE prompt_slug != ''
GROUP BY prompt_slug, prompt_version
ORDER BY prompt_slug, prompt_version

Set the time window with the query's period filter rather than in the SQL itself. Run these from the Query dashboard, the SDK with query.execute(), or the REST API. llm_metrics also exposes ms_to_first_chunk and tokens_per_second for latency and throughput, plus finish_reason, request_model, cached_read_tokens, reasoning_tokens, and per-direction input_cost / output_cost for finer breakdowns.

Next steps