Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

Okoth67/wekeza-llm-finetuning

Domaine:

natural language processingsocioeconomic

Type de record:

modelsoftware
Créateur:
Oko
Hôte:
Wekeza LLM is an open-source financial assistant focused on the Kenyan investment ecosystem. This project fine-tunes the Mistral-7B-Instruct model using QLoRA on a custom Alpaca-style dataset to help users answer questions about money market funds and other investment options in Kenya. # Wekeza LLM: Fine-Tuning Mistral-7B for Kenyan Investment Advice 🇰🇪💰 Wekeza LLM is a language model fine-tuned to provide intelligent answers about money market funds and investment options in Kenya. ## 🚀 Project Goal Build a specialized financial assistant using QLoRA and instruction fine-tuning techniques, designed to: - Understand investment-related questions in Kenya. - Answer in a helpful, local, and practical tone. - Support formats like Alpaca and ChatML for extensibility. ## 🛠️ Tech Stack - Model: `mistralai/Mistral-7B-Instruct-v0.1` - Fine-tuning: QLoRA (via `peft`) - Tokenization: `transformers` library - Dataset: Custom curated in Alpaca format (JSONL)

Visit

github.com

Similaires

Okoth67/distilgpt2-wekeza-finetuned-v3-loraOkoth67/distilgpt2-lora-finetune-wekeza-v3Abdennour03/Darija-LLM-Finetuningbenbel376/llm-finetuning-for-amharicNaomiMeseret/Amharic-LLM-finetuning-pipelineYosef-Ali/amharic-llm-finetuning

Okoth67/distilgpt2-wekeza-finetuned-v3-lora

Fine-tuned DistilGPT2 model using LoRA to generate Kenya-specific financial and investment advice, p

Okoth67/distilgpt2-lora-finetune-wekeza-v3

lgpt2 using backtranslated instructions for Wekeza LLM v3, a Kenya-focused investment assistant #

Abdennour03/Darija-LLM-Finetuning

# XDarija: Fine-Tuning LLMs on Darija & Tamazight This repository is dedicated to fine-tuning lar

benbel376/llm-finetuning-for-amharic

# Amharic RAG Ad Builder ## Project Overview The project is organized into several tasks to achieve

NaomiMeseret/Amharic-LLM-finetuning-pipeline

# Amharic LLM Fine-Tuning Fine-tuning a Large Language Model for Amharic — a low-resource Semitic l

Yosef-Ali/amharic-llm-finetuning

amharic-llm-finetuning # 🇪🇹 Amharic H-Net: Advanced AI Text Generation System A state-of-th