Skip to content

Nexa-gauge

AI Coding Tools

A Python package and CLI toolkit that provides a graph‑based evaluation framework for LLM, RAG, and agentic system outputs, offering repeatable metrics, cost estimation, caching, and structured reports.

Python Latest v0.1.19 · 21d ago Security brief →

Features

  • Graph‑based evaluation pipeline with deterministic node topology
  • Cost‑estimation before execution to preview billable LLM usage
  • Cache‑aware runs that reuse cached artifacts and avoid duplicate spend
  • Structured JSON per‑case reports for CI, dashboards, and analytics

Recent releases

View all 17 releases →
No immediate action
v0.1.19 New feature

multi‑claim extraction

No immediate action
v0.1.18 Breaking risk

refalign node + refmatch rename

No immediate action
v0.1.17 New feature

per‑node scoring contract

No immediate action
v0.1.16 New feature

GEval scoring modes

No immediate action
v0.1.15 Breaking risk

Canonical field name changes

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
40
Forks
13
Languages
Python Makefile Shell

Install & Platforms

Install via
pip

Beta — feedback welcome: [email protected]