Catalogue/Dev tools/HILT

HILT

Trace your LLM calls in a single line of code.

An open-source, privacy-first format for logging human-AI interactions. HILT automatically captures every prompt and response, with latency, tokens and cost, without ever letting your data leave your environment.

View the repoGitHubpip install hilt-pythonPyPI
v0.2.8Version
Python 3.10+Runtime
Apache 2.0License
Open sourceStatus
Inside this edition

Observability, with no compromise on privacy.

No.01

Privacy-first

Your data stays in your environment. No third-party service, GDPR compliance by default.

No.02

One-line setup

A single call to instrument() is enough to automatically trace all your LLM calls.

No.03

Auto-instrumentation

Automatic support for the OpenAI SDK. Anthropic and Gemini are on the roadmap.

No.04

Flexible backends

Local JSONL export for Pandas or Spark, or real-time Google Sheets for support and QA.

Why HILT?

As soon as an application calls a language model, a question arises: what actually happened? Which prompt, which response, what cost, what latency? Existing observability tools answer this, but often at a cost: your data flows through a third-party service.

HILT takes the opposite stance. Logging happens on your side, in local JSONL or in a Google Sheet you control. A single line of code captures every exchange, with latency, tokens, cost and status codes. Apache 2.0 licensed, built for GDPR compliance, and designed to extend to every SDK.

In brief
What is HILT?

HILT is an open-source, privacy-first Python library for logging LLM calls. A single line of code automatically instruments the OpenAI SDK and records every prompt and response with latency, tokens, cost and status codes. Data stays local as JSONL or in a Google Sheet you control. Apache 2.0 licensed, built for GDPR compliance.

Frequently asked

Everything people ask us.

Is HILT a Langfuse, Helicone or LangSmith alternative?
Yes. HILT is an open-source Python library for LLM logging and observability, the same niche as Langfuse, Helicone or LangSmith. Differentiators: privacy-first, local JSONL or Google Sheets backend, GDPR-compliant out of the box, and one-line install.
Which models does HILT support?
HILT auto-instruments the OpenAI SDK (GPT-4, GPT-4o, GPT-5 and later) and is extensible to other SDKs.
Discover next

Other editions from the catalogue.

No.11MakoReal-time data pipelines, in YAML.No.12VectLiteAn embedded vector database for local AI.No.13Données Québec MCPAn MCP server for Quebec's open data.

Instrument your AI agents.

Open source, Apache 2.0 licensed. Install HILT in one line and keep control of your data.

View the repoGitHubpip install hilt-pythonPyPI

In brief

How does HILT log LLM calls?

HILT (Human-AI Log Tracing) is an open-source Python library that auto-instruments OpenAI SDK calls with a single import line. Once enabled, HILT captures 11 columns per request: timestamp, model, latency in milliseconds, input and output token counts, estimated cost in USD, user prompt, model response, session identifier, custom metadata, and success or error status. Logs are written to a local JSONL file by default, or pushed to a Google Sheets spreadsheet if the user configures a service account key. HILT supports GPT-3.5, GPT-4, GPT-4o and GPT-5 models from day one, plus any model compatible with the standard OpenAI API. The library is under 800 lines of pure Python code.

Is HILT GDPR-compliant?

HILT is designed around 3 documented GDPR principles. First, local storage by default: no data leaves the application's machine without explicit configuration, which simplifies the legal basis for processing. Second, automatic redaction hooks mask emails, phone numbers, card numbers, and IP addresses via regular expressions provided in the documentation. Third, every JSONL line contains a `redacted` field describing the transformations applied, which streamlines audits. HILT is distributed under the Apache 2.0 license on PyPI at https://pypi.org/project/hilt-python/ and the source code is available on GitHub at https://github.com/Stefen-Taime/hilt-python. The library is maintained by mcsÉdition and used in production by engineering teams of fewer than 50 people.

Frequently asked questions

Is HILT a Langfuse, Helicone or LangSmith alternative?

Yes. HILT is an open-source Python library for LLM logging and observability — same niche as Langfuse, Helicone or LangSmith. Differentiators: privacy-first, local JSONL or Google Sheets backend, GDPR-compliant out of the box, and one-line install.

Which models does HILT support?

HILT auto-instruments the OpenAI SDK (GPT-4, GPT-4o, GPT-5...) and is extensible to other SDKs.

HILT — Human-AI Log Trace, Privacy-First LLM Logging | mcsÉdition