PHUDGE logo

PHUDGE

deshwalmahesh/PHUDGE

Evaluation toolkit for LLM responses with scalable grading and hallucination detection.

GraphCanon updated Sep 10, 2026 · GitHub synced Sep 10, 2026

22views this month

53 stars7 forksLast push Jul 10, 2024 Jupyter Notebook

Decision brief

PHUDGE grades LLM responses using scalable configurations with custom rubrics and reference answers for both relative and absolute grading.

Good fit when

  • Need robust evaluation of LLM responses on a scale from one to five
  • Require scalable hallucination detection in LLM outputs

Avoid when

  • Looking for real-time deployment without setup for different environments
  • Need support for models not compatible with Phi-3 architecture
Pricing:
freemium - Primary model is free; potential costs associated with training and running on specific hardware.

Observed Jul 16, 2026 · Source: enrich:decision_facts

Verify the decision

Maintenance and security

Full trust report
Maintenance
Dormant (791d since push)
As of Sep 10, 2026
Provenance
Not a fork · Personal account
As of Sep 10, 2026
Security (OSV)
No lockfile
As of Jul 15, 2026

Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.

Install

git clone https://github.com/deshwalmahesh/PHUDGE

Similar tools

Same-category neighbours. No typed graph edges are catalogued for this tool yet.

Evidence and technical details

Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.

Overview

PHUDGE is a model designed to evaluate language model responses on a scale from one to five using various configurations including custom rubrics and reference answers, focusing on both relative and absolute grading. The project offers tools for training, evaluation, and deployment of this model.

Capability facts

Languages
jupyter notebook

Source: github.language · Sep 10, 2026

Categories

Compatibility

Sourced claims from the README excerpt - not unsourced marketing copy.

Python runtimePython

Source: README excerpt (regex_v1, Sep 10, 2026)

e, Phi-3 will be available in the main version. This code works perfectly with `Python 3.10 + A10 GPU`. Look for the version you are using. I repeat, use the same versions
Source link

Tags

README

PHUDG3: Phi 3 as Scalable Judge is for The Legend himself Paper Link What is it? A robust production grade and scalable SOTA (4 Benchmarks) model for Relative and Absolute grading of LLM (as well human) responses. What it can do? Given a question and it's response, it can judge t...

For agents

This page has a .md twin and JSON over the API.

Was this helpful?

Anonymous feedback helps us improve pages and translations.