Jore

C 68 completed
Ai Ml
unknown / text · small
94
Files
18,712
LOC
1
Frameworks
9
Languages

Pipeline State

completed
Run ID
#366470
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Cataloged
Decision
proceed
Novelty
61.33
Framework unique
Isolation
Last stage change
2026-05-10 03:35:17
Deduplication group #52529
Member of a group with 5 similar repo(s) — canonical #90648 view group →
Top concepts (2)
Project DescriptionWeb Backend
About: code-quality intelligence by Repobility · https://repobility.com

AI Prompt

Create a small, self-contained application based on the Jore nano transformer LLM. I need a Flask web UI that allows users to interact with the model for natural language processing tasks. The UI should support both a chat interface and a quiz functionality. Since the core inference engine is in C99, please ensure the Flask backend can manage the model loading and run inference, perhaps using the provided `inference/` directory structure. Also, include the necessary setup for training, referencing the PyTorch training loop and the cron automation for periodic evaluation.
python flask llm pytorch nlp c99 web-ui transformer text-generation
Generated by gemma4:latest

Catalog Information

jore is a nano transformer LLM built from scratch, designed for natural language processing tasks.

Description

jore is an open-source, PyTorch-based transformer model that can be trained and used for various NLP tasks. It features a C99 inference engine, a Flask web UI, and supports char-level tokenization. The project includes a training loop, checkpoint management, and overnight automation using cron.

الوصف

يور هو نموذج معالجة اللغة الطبيعية من الدرجة النانوية، مصمم لتحليل اللغة الطبيعية. يعتمد على بيرش ويتش (PyTorch) ويعمل على تشغيل المعالجات المحلية (C99). يحتوي على واجهة ويب باستخدام فلاسك (Flask)، وتدعم التكامل بالحروف. يتضمن المشروع حلقة تدريب، إدارة نقاط التحقق، وتسريع الليل باستخدام الكرون.

Novelty

7/10

Tags

natural-language-processing transformer-model char-level-tokenization overnight-automation cron flask-web-ui

Technologies

flask matplotlib numpy pytorch

Claude Models

claude-sonnet-4.5 claude-opus-4.6

Quality Score

C
68.5/100
Structure
76
Code Quality
74
Documentation
58
Testing
75
Practices
53
Security
70
Dependencies
60

Strengths

  • CI/CD pipeline configured (github_actions)
  • Good test coverage (48% test-to-source ratio)
  • Consistent naming conventions (snake_case)
  • Properly licensed project

Weaknesses

  • 243 duplicate lines detected \u2014 consider DRY refactoring

Recommendations

  • Add a linter configuration to enforce code style consistency

Security & Health

6.1h
Tech Debt (A)
A
OWASP (100%)
PASS
Quality Gate
A
Risk (1)
Repobility · MCP-ready · https://repobility.com
Apache-2.0
License
19.6%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

text
60.7%
python
28.8%
html
4.9%
c
2.1%
markdown
1.6%
shell
1.3%
json
0.3%
yaml
0.2%
ini
0.0%

Frameworks

Flask

Concepts (2)

Findings produced by Repobility · scan your repo at https://repobility.com/scan/
CategoryNameDescriptionConfidence
Repobility · code-quality intelligence platform · https://repobility.com
auto_descriptionProject DescriptionNano transformer LLM built from scratch. PyTorch training loop, C99 inference engine, overnight automation.80%
auto_categoryWeb Backendweb-backend70%

Quality Timeline

1 quality score recorded.

View File Metrics

Embed Badge

Add to your README:

![Quality](https://repos.aljefra.com/badge/90653.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV