Command-Line Multi-GPU Fine-Tuning of Large Language Models Using LLaMA-Factory

Model Preparation There are several reliable methods to download pre-trained models: ModelScope (recommended for fast download speeds and includes many restricted models) Hugging Face mirror sites (requires model access permisions) Public cloud storage resources (use tools like XShell for faster uploads) Dataset Preparation Two primary datase ...

Posted on Sat, 19 Sep 2026 16:11:57 +0000 by php-coder

Optimizing Qwen2.5-VL: Supervised Fine-Tuning and Preference Alignment Workflows

Input Templating and Vision Tokenization The foundational step in deploying Qwen2.5-VL involves structuring multimodal inputs into a standardized conversational schema. The tokenizer expects a sequence of role-annotated exchanges containing both textual instructions and visual payloads. Below is a refactored approach to constructing these paylo ...

Posted on Thu, 17 Sep 2026 16:53:27 +0000 by simplyi

Fine-Tuning and Deployment of Llama3 with Limited GPU Memory

When fine-tuning large language models on consumer-grade GPUs with limited VRAM—such as an RTX 4060 Ti with 16GB—memory constraints become a major bottleneck. This article outlines a practical workflow for fine-tuning and deploying the Llama3-Chinese-8B-Instruct model under these conditions. Initial Attempt with LLaMA-Factory The first appro ...

Posted on Tue, 25 Aug 2026 16:41:52 +0000 by vishwavivek

Advanced NL2SQL: Production-Ready Open-Source Stacks with DB-GPT-Hub, SQLCoder, and Text-to-SQL Fine-Tuning

MindSQL – One-Line RAG for Database Chat MindSQL is a lightweight Python library that turns any relational database into a conversational agent. It plugs into PostgreSQL, MySQL, SQLite, Snowflake, BigQuery, and more, while letting you pick any LLM (GPT-4, Llama-2, Gemini) and vector store (Chroma, FAISS). pip install mindsql from mindsql.co ...

Posted on Sat, 08 Aug 2026 16:56:46 +0000 by The Saint