×
Please Note! This page is still a work in progress.
Large Languagle Models (LLMs) & Natural Language Processing (NLP)
Est. read time: 2 minutes | Last updated: July 13, 2025 by John Gentile
Contents
Overview
Large Language Models (LLM)
Popular Cloud LLMs
- ChatGPT - OpenAI
- Claude - Anthropic
- Gemini - Google, which also has other tools like AI Studio which also has a very large context window.
- Grok - xAI
LLMs for Coding
- LLMs have varied abilities when it comes to coding.
- (3/27/2025) so far Claude 3.7 Sonnet has slightly better coding results.
- Various vendor and open-source tools exist like:
- Claude Code
- yamadashy/repomix: packs an entire repository into a single, AI-friendly file, which can then be uploaded to most LLMs.
LLMs for Research and Knowledge
- Claude, OpenAI/ChatGPT and Grok have “research” modes where the model is open to searching the internet, and has multi-step reasoning to come up with deep assessments and recommendations.
- As of 7/12/2025, Grok 4 seems to be the SOTA model for knowledge and research.
- Google’s NotebookLM is targeted at helping research in Jupyter-style notebooks.
LLMs for Image Generation
- ChatGPT’s Sora is great for realistic and photo image generation.
- Most general models can synthesize an image from a prompt if the result is a text-based image format such as SVG, Mermaid, HTML/CSS, etc.
Prompt Engineering
Suggestions:
- Start with a short and simple prompt, and iterate to get better results.
- Put instructions at beginning or end of prompt, while clearly separating instructions from the text of interest.
- Describe how the model should behave and respond- for example, if looking for coding advice, can create a system prompt of
You are a helpful assistant that answers programming questions.
- Add specificity and descriptions about the task and desired output, as well as including examples of output if possible.
- Instructions should tell what “to do” rather than “what not to do”.
References:
- Prompt Engineering Guide
- Prompt engineering - Hugging Face
- Prompt engineering - OpenAI
- OpenAI Cookbbok
- Anthropic’s Prompt Engineering Interactive Tutorial
Self-Hosted
- Ollama: easily run LLMs locally. Very easy to setup and start.
- llama.cpp: fast, low overhead inference of LLMs in C/C++ that runs under-the-hood of Ollama.
- Hugging Face: pre-trained NLP models & reference
- LangChain: LangChain is a Python framework for developing applications powered by large language models (LLMs).
Other Uses/Automation
- LaurieWired/GhidraMCP: ghidraMCP is an Model Context Protocol server for allowing LLMs to autonomously reverse engineer applications. It exposes numerous tools from core Ghidra functionality to MCP clients.
Architecture & References
- Attention Is All You Need- Arxiv: introduces concepts of transformers and attention
- LLaMA: Open and Efficient Foundation Language Models - arXiv: Meta AI open-source LLM model.
- llama3 implemented from scratch
- Transformer Inference Arithmetic
- Transformer Math 101 with a focus on training cost.
- Training data-efficient image transformers & distillation through attention- Facebook AI
- The Illustrated Transform: NLP walk-through
- Tiny LLM - LLM Serving in a Week
- DeepSeek v3 & R1 Model Architecture
- How to Scale Your Model- A Systems View of LLMs on TPUs - Jax-ML