Hugging Face Observability & Monitoring with OpenTelemetry

SigNoz released a new integration guide for monitoring Hugging Face AI applications using OpenTelemetry, enabling developers to export traces, logs, and metrics to the SigNoz observability platform. The guide provides step-by-step instructions for setting up automatic instrumentation in Python environments, allowing teams to track LLM usage and application performance through unified dashboards and alerts. This integration aims to improve reliability and user experience for AI workflows by providing correlated telemetry data in a single monitoring interface.

Overview This guide walks you through setting up observability and monitoring for Hugging Face using OpenTelemetry https://opentelemetry.io/ and exporting traces, logs, and metrics to SigNoz. With this integration, you can observe and track various metrics for your Hugging Face applications and LLM usage. Monitoring Hugging Face in your AI applications with telemetry ensures full observability across your AI and LLM workflows. By leveraging SigNoz, you can analyze correlated traces, logs, and metrics in unified dashboards, configure alerts, and gain actionable insights to continuously improve reliability, responsiveness, and user experience. Prerequisites - A SigNoz Cloud account https://signoz.io/teams/ with an active ingestion key or Self Hosted SigNoz instance https://signoz.io/docs/install/self-host/ - For Python: pip installed for managing Python packages - A Hugging Face Account and API key. You can get the API key from Hugging Face platform https://huggingface.co/settings/tokens Monitoring Hugging Face For more information on getting started with Hugging Face in your Python environment, refer to the Hugging Face quickstart guide https://huggingface.co/inference/get-started . No code auto-instrumentation is recommended for quick setup with minimal code changes. It's ideal when you want to get observability up and running without modifying your application code and are leveraging standard instrumentor libraries. Step 1: Install the necessary packages in your Python environment. pip install \ opentelemetry-distro \ opentelemetry-exporter-otlp \ httpx \ opentelemetry-instrumentation-httpx \ opentelemetry-instrumentation-system-metrics \ openai \ opentelemetry-instrumentation-openai-v2 Step 2: Add Automatic Instrumentation opentelemetry-bootstrap --action=install Step 3: Configure logging level To ensure logs are properly captured and exported, configure the root logger to emit logs at the DEBUG level or higher: python import logging logging.getLogger .setLevel logging.DEBUG logging.getLogger "httpx" .setLevel logging.DEBUG This sets the minimum log level for the root logger to DEBUG, which ensures that logger.debug calls and higher severity logs INFO, WARNING, ERROR, CRITICAL are captured by the OpenTelemetry logging auto-instrumentation and sent to SigNoz. Step 4: Create an example Hugging Face application python import os from openai import OpenAI client = OpenAI base url="https://router.huggingface.co/v1", api key=os.environ "HF TOKEN" , completion = client.chat.completions.create model="moonshotai/Kimi-K2-Instruct-0905", messages= { "role": "user", "content": "What is SigNoz?", } , print completion.choices 0 .message Before running this code, ensure that you have set the environment variable HF TOKEN with your generated Hugging Face API key. Step 5: Run your application with auto-instrumentation Run your application with the following environment variables set. This configures OpenTelemetry to export traces, logs, and metrics to SigNoz Cloud and enables automatic log correlation: OTEL RESOURCE ATTRIBUTES="service.name=<service name " \ OTEL EXPORTER OTLP ENDPOINT="https://ingest.<region .signoz.cloud:443" \ OTEL EXPORTER OTLP HEADERS="signoz-ingestion-key=<your-ingestion-key " \ OTEL EXPORTER OTLP PROTOCOL=grpc \ OTEL TRACES EXPORTER=otlp \ OTEL METRICS EXPORTER=otlp \ OTEL LOGS EXPORTER=otlp \ OTEL PYTHON LOG CORRELATION=true \ OTEL PYTHON LOGGING AUTO INSTRUMENTATION ENABLED=true \ opentelemetry-instrument <your run command is the name of your service <service name <region : Your SigNoz Cloud region https://signoz.io/docs/ingestion/signoz-cloud/overview/ endpoint <your-ingestion-key : Your SigNoz ingestion key https://signoz.io/docs/ingestion/signoz-cloud/keys/ - Replace <your run command with the actual command you would use to run your application. In this case we would use: python main.py Code-based manual instrumentation gives you fine-grained control over your telemetry configuration. Use this approach when you need to customize resource attributes, sampling strategies, or integrate with existing observability infrastructure. Step 1: Install additional OpenTelemetry dependencies pip install \ opentelemetry-api \ opentelemetry-sdk \ opentelemetry-exporter-otlp \ opentelemetry-instrumentation-httpx \ opentelemetry-instrumentation-system-metrics \ openai \ opentelemetry-instrumentation-openai-v2 Step 2: Set up Traces python from opentelemetry.sdk.resources import Resource from opentelemetry.sdk.trace import TracerProvider from opentelemetry.sdk.trace.export import BatchSpanProcessor from opentelemetry.exporter.otlp.proto.http.trace exporter import OTLPSpanExporter from opentelemetry import trace from opentelemetry.instrumentation.openai v2 import OpenAIInstrumentor import os resource = Resource.create {"service.name": "<service name "} provider = TracerProvider resource=resource span exporter = OTLPSpanExporter endpoint= os.getenv "OTEL EXPORTER TRACES ENDPOINT" , headers={"signoz-ingestion-key": os.getenv "SIGNOZ INGESTION KEY" }, processor = BatchSpanProcessor span exporter provider.add span processor processor trace.set tracer provider provider OpenAIInstrumentor .instrument is the name of your service <service name → SigNoz Cloud trace endpoint with appropriate OTEL EXPORTER TRACES ENDPOINT region https://signoz.io/docs/ingestion/signoz-cloud/overview/ endpoint : https://ingest.<region .signoz.cloud:443/v1/traces → Your SigNoz SIGNOZ INGESTION KEY ingestion key https://signoz.io/docs/ingestion/signoz-cloud/keys/ Step 3: Set up Metrics python from opentelemetry.sdk.resources import Resource from opentelemetry.sdk.metrics import MeterProvider from opentelemetry.exporter.otlp.proto.http.metric exporter import OTLPMetricExporter from opentelemetry.sdk.metrics.export import PeriodicExportingMetricReader from opentelemetry import metrics from opentelemetry.instrumentation.system metrics import SystemMetricsInstrumentor import os resource = Resource.create {"service.name": "<service name "} metric exporter = OTLPMetricExporter endpoint= os.getenv "OTEL EXPORTER METRICS ENDPOINT" , headers={"signoz-ingestion-key": os.getenv "SIGNOZ INGESTION KEY" }, reader = PeriodicExportingMetricReader metric exporter metric provider = MeterProvider metric readers= reader , resource=resource metrics.set meter provider metric provider meter = metrics.get meter name turn on out-of-the-box metrics SystemMetricsInstrumentor .instrument HTTPXClientInstrumentor .instrument is the name of your service <service name → SigNoz Cloud endpoint with appropriate OTEL EXPORTER METRICS ENDPOINT region https://signoz.io/docs/ingestion/signoz-cloud/overview/ endpoint : https://ingest.<region .signoz.cloud:443/v1/metrics → Your SigNoz SIGNOZ INGESTION KEY ingestion key https://signoz.io/docs/ingestion/signoz-cloud/keys/ SystemMetricsInstrumentor provides system metrics CPU, memory, etc. , and HTTPXClientInstrumentor provides outbound HTTP request metrics such as request duration. If you want to add custom metrics to your Hugging Face application, see Python Custom Metrics https://signoz.io/opentelemetry/python-custom-metrics/ . Step 4: Set up Logs python import logging from opentelemetry.sdk.resources import Resource from opentelemetry. logs import set logger provider from opentelemetry.sdk. logs import LoggerProvider, LoggingHandler from opentelemetry.sdk. logs.export import BatchLogRecordProcessor from opentelemetry.exporter.otlp.proto.http. log exporter import OTLPLogExporter import os resource = Resource.create {"service.name": "<service name "} logger provider = LoggerProvider resource=resource set logger provider logger provider otlp log exporter = OTLPLogExporter endpoint= os.getenv "OTEL EXPORTER LOGS ENDPOINT" , headers={"signoz-ingestion-key": os.getenv "SIGNOZ INGESTION KEY" }, logger provider.add log record processor BatchLogRecordProcessor otlp log exporter Attach OTel logging handler to root logger handler = LoggingHandler level=logging.INFO, logger provider=logger provider logging.basicConfig level=logging.INFO, handlers= handler logger = logging.getLogger name Enable httpx logging to capture HTTP requests httpx logger = logging.getLogger "httpx" httpx logger.setLevel logging.DEBUG httpx logger.addHandler handler is the name of your service <service name → SigNoz Cloud endpoint with appropriate OTEL EXPORTER LOGS ENDPOINT region https://signoz.io/docs/ingestion/signoz-cloud/overview/ endpoint : https://ingest.<region .signoz.cloud:443/v1/logs → Your SigNoz SIGNOZ INGESTION KEY ingestion key https://signoz.io/docs/ingestion/signoz-cloud/keys/ Step 5: Run an example Hugging Face application Ensure you have completed the steps above traces, logs, and metrics configuration before running this code. All OpenTelemetry instrumentation must be initialized first. python import os from openai import OpenAI client = OpenAI base url="https://router.huggingface.co/v1", api key=os.environ "HF TOKEN" , completion = client.chat.completions.create model="moonshotai/Kimi-K2-Instruct-0905", messages= { "role": "user", "content": "What is SigNoz?", } , print completion.choices 0 .message Before running this code, ensure that you have set the environment variable HF TOKEN with your generated Hugging Face API key. View Traces, Logs, and Metrics in SigNoz Your Hugging Face LLM usage should now automatically emit traces, logs, and metrics. You should be able to view traces in Signoz Cloud under the traces tab: When you click on a trace in SigNoz, you'll see a detailed view of the trace, including all associated spans, along with their events and attributes. You should be able to view logs in SigNoz Cloud under the logs tab. You can also view logs by clicking on the “Related Logs” button in the trace view to see correlated logs: When you click on any of these logs in SigNoz, you'll see a detailed view of the log, including attributes: You should be able to see Hugging Face related metrics in Signoz Cloud under the metrics tab: When you click on any of these metrics in SigNoz, you'll see a detailed view of the metric, including attributes: Troubleshooting Troubleshooting troubleshooting If you don't see your telemetry data: Verify network connectivity - Ensure your application can reach SigNoz Cloud endpoints Check ingestion key - Verify your SigNoz ingestion key is correct Wait for data - OpenTelemetry batches data before sending, so wait 10-30 seconds after making API calls Try a console exporter — Enable a console exporter locally to confirm that your application is generating telemetry data before it’s sent to SigNoz Next Steps You can also check out our custom Hugging Face dashboard here https://signoz.io/docs/dashboards/dashboard-templates/huggingface-dashboard/ which provides specialized visualizations for monitoring your Hugging Face usage in applications. The dashboard includes pre-built charts specifically tailored for LLM usage, along with import instructions to get started quickly. Setup OpenTelemetry Collector Optional Setup OpenTelemetry Collector Optional setup-opentelemetry-collector-optional What is the OpenTelemetry Collector? Think of the OTel Collector as a middleman between your app and SigNoz. Instead of your application sending data directly to SigNoz, it sends everything to the Collector first, which then forwards it along. Why use it? Cleaning up data — Filter out noisy traces you don't care about, or remove sensitive info before it leaves your servers. Keeping your app lightweight — Let the Collector handle batching, retries, and compression instead of your application code. Adding context automatically — The Collector can tag your data with useful info like which Kubernetes pod or cloud region it came from. Future flexibility — Want to send data to multiple backends later? The Collector makes that easy without changing your app. See Switch from direct export to Collector https://signoz.io/docs/opentelemetry-collection-agents/opentelemetry-collector/switch-to-collector/ for step-by-step instructions to convert your setup. For more details, see Why use the OpenTelemetry Collector? https://signoz.io/docs/opentelemetry-collection-agents/opentelemetry-collector/why-to-use-collector/ and the Collector configuration guide https://signoz.io/docs/opentelemetry-collection-agents/opentelemetry-collector/configuration/ . Additional resources: - Set up alerts https://signoz.io/docs/alerts/ for high latency or error rates - Learn more about querying traces https://signoz.io/docs/userguide/traces/ - Explore log correlation https://signoz.io/docs/userguide/logs query builder/