Government Of India
A-
A
A+
COMMUNITY 
Upvote 0
RAG Engine
Python script that implements a Retrieval‑Augmented Generation (RAG) pipeline for document question answering.
- 0upvotes
0downloads
3views
0
0Shares
updated 05 Nov 2026
Upvote 0
TASK Text Generation
LICENSE Apache 2.0
LANGUAGE English
FORMAT Other
SECTOR Other
About Model
The repository provides a single Python file that defines a Retrieval‑Augmented Generation (RAG) engine suitable for integration into AIKosh pipelines. The core components are a SemanticRetriever class, which loads pre‑computed document embeddings and uses cosine similarity (via scikit‑learn) to find the most relevant document chunks for a user query, and a RAGPipeline class that combines this retriever with a language model pipeline to generate answers based on retrieved context. The pipeline accepts a textual query, returns a dictionary containing the generated answer and the supporting documents with relevance scores, and allows configuration of retrieval depth, token limits, temperature, and context window size. The script requires external embedding and language model pipelines; it does not contain model weights or a trained model itself. It is intended for developers building custom document‑based question‑answering applications and can be extended with different embedding or LLM back‑ends.
Intended Usecase
The script can be used to build internal knowledge‑base assistants, FAQ chatbots, and customer‑support tools that answer user questions from a collection of documents. It is suitable for enterprises that have pre‑computed document embeddings and want to add contextual language model responses without building a full RAG system from scratch. Developers can also adapt it for research prototypes exploring retrieval‑augmented generation techniques.
Limitations
The repository contains only a single Python file and no pre‑computed embeddings or language model weights, so it cannot be run as a stand‑alone service. It relies on external embedding and LLM pipelines supplied by the user, and assumes that document embeddings are compatible with cosine similarity. The code does not include error handling for mismatched dimensions, large document collections, or latency‑critical deployments. It is not a complete end‑to‑end solution and is limited to text‑based retrieval and generation.
Metadata
Apache 2.0
Other
NAA
NA
Text Generation
Other
Transformers
Gemma
20
NA
English
Open
05/11/26 19:37:35
Hosted
0
Tags
Related Models
tokenmodel
How people read your model before they download it.
Joel-Speech-TTS
A Text-to-Speech (TTS) model developed for the Mizo language that converts written Mizo text into natural-sounding speech, supporting the development of Mizo-language voice applications and accessibility solutions.
mizo-f5-tts-v1
A Text-to-Speech (TTS) model developed for the Mizo language that converts written Mizo text into natural-sounding speech, supporting the development of Mizo-language voice applications and accessibility solutions.
CoRover-ASR-hindi-conformer-hybrid-large
ASR-Hindi-Conformer-Hybrid-Large is a Hindi automatic speech recognition model developed by CoRover, based on the FastConformer architecture with hybrid RNNT and CTC decoding for accurate speech-to-text transcription.
CoRover-BharatGPT-Instruct
BharatGPT-Instruct E2B is a multilingual language model by CoRover.ai, designed to deliver high-quality, culturally relevant AI responses across Indian languages.
CoRover-BharatGPT-mini
BharatGPT Mini - A compact, Transformer-based language model developed by CoRover, pretrained on large-scale text data to understand and generate natural language, advancing India’s indigenous AI capabilities.
CoRover-BharatGPT-3B-Indic
A multilingual model fine-tuned on authentic conversational data across Indic languages, supporting translation, summarisation, and text generation. Ideal for Secure RAG (S-RAG) and domain-specific fine-tuning, it enables the development of sovereign, multilingual AI agents, chatbots, voicebots, and virtual assistants.
Samadhan-Bharat-Epoch1-Tri-Statutory-Neural-Model
Sovereign 5-Head Tri-Statutory Deep Neural Engine integrating 22 Indic constitutional languages, MSMED Act Section 16 monthly compounding (3x RBI rate), CGST Section 16(2)(b) statutory delivery estoppel, Section 43B(h) Income Tax profit disallowances, and RBI CRILC delinquency classification.
Samadhan-Bharat-Sovereign-Indic-Voice-RVC-Acoustic-Engine
Deep neural acoustic adjudication and Retrieval-Based Voice Conversion (RVC) engine trained on IndiaAI cloud infrastructure. Features 1,976,327 trainable parameters with a 4-layer LayerNorm SiLU backbone and multi-head statutory predictors. Simultaneously predicts invoice default principal, statutory delay horizon, Section 16 monthly compound interest (at 3x RBI Bank Rate), and 4-quadrant legal classification from 256-D vernacular speech feature tensors across India's 22 scheduled languages.
Related Models by Organization
tokenmodel
How people read your model before they download it.
Samadhan-Bharat-Epoch1-Tri-Statutory-Neural-Model
Sovereign 5-Head Tri-Statutory Deep Neural Engine integrating 22 Indic constitutional languages, MSMED Act Section 16 monthly compounding (3x RBI rate), CGST Section 16(2)(b) statutory delivery estoppel, Section 43B(h) Income Tax profit disallowances, and RBI CRILC delinquency classification.
Samadhan-Bharat-Sovereign-Indic-Voice-RVC-Acoustic-Engine
Deep neural acoustic adjudication and Retrieval-Based Voice Conversion (RVC) engine trained on IndiaAI cloud infrastructure. Features 1,976,327 trainable parameters with a 4-layer LayerNorm SiLU backbone and multi-head statutory predictors. Simultaneously predicts invoice default principal, statutory delay horizon, Section 16 monthly compound interest (at 3x RBI Bank Rate), and 4-quadrant legal classification from 256-D vernacular speech feature tensors across India's 22 scheduled languages.
Samadhan-Bharat-Sovereign-4Quadrant-2B-Master-Engine
Sovereign 4-Quadrant Dual-Pincer neural engine for simultaneous MSME delayed trade liquidity unfreezing and GST circular trading fraud detection, trained and validated over 1.35 Trillion statutory transactions.
Samadhan-Bharat-Sovereign-Dual-Head-Model
Dual-head sovereign legal AI model trained on IndiaAI AIRAWAT and 16-Core CPU supercluster across 15.7M+ tokens. Features a 95.04% accurate Judicial Adjudication Head for Section 18 MSMED disputes and a 100.00% accurate Treasury Anti-Fraud Head for circular GST invoice interception.
Test model
Model Short Description
NCERT-TUTOR-v3
NCERT-aligned tutor LoRA for on-device K-12 tutoring. Runs on 4 GB VRAM (RTX 3050).
Stage-3 chained adapter: 15.7k curated questions (Grades 6-12, 13 subjects) + 3 DPO stages
(math reasoning, mined factuality, multi-subject). 58% on held-out NCERT math with worked steps.
Kaveri-4B
Kaveri-4B is a multilingual text-generation model for conversational AI, instruction following, reasoning, coding assistance, summarization, translation, and general-purpose NLP tasks.
Kaveri-1-5B
Kaveri-1.5B is a lightweight multilingual text-generation assistant by RiyaVibe for general chat, instruction following, basic question answering, simple coding help, basic math, translation, and local or cloud inference.
Related Models by Sector
Joel-Speech-TTS
A Text-to-Speech (TTS) model developed for the Mizo language that converts written Mizo text into natural-sounding speech, supporting the development of Mizo-language voice applications and accessibility solutions.
mizo-f5-tts-v1
A Text-to-Speech (TTS) model developed for the Mizo language that converts written Mizo text into natural-sounding speech, supporting the development of Mizo-language voice applications and accessibility solutions.
CoRover-ASR-hindi-conformer-hybrid-large
ASR-Hindi-Conformer-Hybrid-Large is a Hindi automatic speech recognition model developed by CoRover, based on the FastConformer architecture with hybrid RNNT and CTC decoding for accurate speech-to-text transcription.
CoRover-BharatGPT-Instruct
BharatGPT-Instruct E2B is a multilingual language model by CoRover.ai, designed to deliver high-quality, culturally relevant AI responses across Indian languages.
CoRover-BharatGPT-mini
BharatGPT Mini - A compact, Transformer-based language model developed by CoRover, pretrained on large-scale text data to understand and generate natural language, advancing India’s indigenous AI capabilities.
CoRover-BharatGPT-3B-Indic
A multilingual model fine-tuned on authentic conversational data across Indic languages, supporting translation, summarisation, and text generation. Ideal for Secure RAG (S-RAG) and domain-specific fine-tuning, it enables the development of sovereign, multilingual AI agents, chatbots, voicebots, and virtual assistants.
Streetlight Working or Not-Working Classification
MobileNetV2-based binary image classification model for automated streetlight visual inspection built using the Aargus DIY visual inspection tool. The model classifies streetlights as Working or Not Working from images or video frames and is intended to support municipal infrastructure monitoring, visual inspection, smart-city applications and maintenance prioritization. The model achieves 95.16% accuracy, 91.30% precision and 95.45% recall on the reported evaluation data.
Welding Defect Detection
A YOLOv8-based computer vision model developed using the Aargus DIY Visual Inspection Tool for automated visual inspection of welds. The model detects and classifies three categories: Bad Weld, Good Weld, and Defect. It is designed to support automated industrial quality inspection and welding defect identification from images.
MIZO LANGUAGE OCR
This dataset is designed for training and evaluating Optical Character Recognition (OCR) models capable of extracting text from document images. It contains 14,000 labeled samples consisting of image filenames and their corresponding ground-truth text annotations, along with metadata such as font type, font size, and document category. The dataset includes diverse document styles, including printed books and scanned documents, enabling models to learn robust text recognition under varying visual
Surface defects classification of the hot-rolled steel strip
A ResNet50-based deep learning model for classifying surface defects
on hot-rolled steel strips using the benchmark dataset.
Achieves 97.50% accuracy across 6 defect categories using transfer
learning and data augmentation techniques.
Related Models by Task Type
CoRover-BharatGPT-Instruct
BharatGPT-Instruct E2B is a multilingual language model by CoRover.ai, designed to deliver high-quality, culturally relevant AI responses across Indian languages.
CoRover-BharatGPT-mini
BharatGPT Mini - A compact, Transformer-based language model developed by CoRover, pretrained on large-scale text data to understand and generate natural language, advancing India’s indigenous AI capabilities.
CoRover-BharatGPT-3B-Indic
A multilingual model fine-tuned on authentic conversational data across Indic languages, supporting translation, summarisation, and text generation. Ideal for Secure RAG (S-RAG) and domain-specific fine-tuning, it enables the development of sovereign, multilingual AI agents, chatbots, voicebots, and virtual assistants.
Kaveri-4B
Kaveri-4B is a multilingual text-generation model for conversational AI, instruction following, reasoning, coding assistance, summarization, translation, and general-purpose NLP tasks.
Kaveri-1-5B
Kaveri-1.5B is a lightweight multilingual text-generation assistant by RiyaVibe for general chat, instruction following, basic question answering, simple coding help, basic math, translation, and local or cloud inference.
Kaveri-05b
Kaveri 0.5B is a lightweight multilingual conversational AI assistant
designed for local inference, instruction following, chat, and agent workflows.
gnani-evon-v3-30B-A3B
Hybrid Mamba2–Transformer MoE, 30B total / ~3.5B active, trained end-to-end for Indic capability across 11 languages. 128K context, tool calling, native-script output. Apache-2.0.
Phi-1.5 - Lightweight Transformer Model for Text Generation and Coding
A 1.3B parameter Transformer model trained on structured QA, NLP tasks, and Python code, optimized for text generation, summarization, and creative writing, with no fine-tuning from human feedback.
Phi-3-Mini-4K-Instruct - Lightweight AI Model for Reasoning and Efficient AI Applications
A 3.8B parameter AI model optimized for memory-efficient reasoning, instruction following, and long-context processing, designed for compute-constrained environments and latency-sensitive applications.
Phi-4-Mini-Instruct - Lightweight Multilingual AI Model
A lightweight AI model optimized for multilingual text generation, reasoning, and instruction adherence, designed for memory-efficient applications and generative AI tasks.
© 2026 - Copyright AIKosh. All rights reserved.