---
title: "Get system and model inference metrics"
method: GET
path: "/v1/metrics"
tags: ["Metrics"]
---

# Get system and model inference metrics

`GET /v1/metrics`

## Request

- Base URL: `http://localhost:9001`
- URL: `http://localhost:9001/v1/metrics`
- Auth: none required

Returns a unified snapshot of live system resource metrics (CPU, memory) and per-model moving averages for all pipeline stages (last 100 requests).

## Response `200`

Success

- MetricsResponse: Unified snapshot of system resource metrics and per-model inference metrics.
  - `system` SystemMetrics, required: Live system resource utilization.
    - `cpu_usage_percent` number, float, required: Current CPU utilization (0–100).
    - `memory_total_gb` number, float, required: Total system memory in GB.
    - `memory_used_gb` number, float, required: Used system memory in GB.
    - `memory_free_gb` number, float, required: Available system memory in GB.
    - `memory_usage_percent` number, float, required: Memory utilization percentage (0–100).
  - `models` object, required: Map of model ID to its moving-average inference metrics. Empty object if no requests have been processed yet.

---

[API](https://skmtc.dev/qualcomm/apis/llm-vlm-genai-microservice-apis.md) - [All operations](https://skmtc.dev/qualcomm/apis/llm-vlm-genai-microservice-apis/llms.txt) - [OpenAPI document](https://skmtc.dev/qualcomm/apis/llm-vlm-genai-microservice-apis/revisions/8557fe985a5e?raw)
