LLM-handledningar

Setting up llama.cpp to run the GLM-4.7 model on a single NVIDIA H100 80GB GPU, achieving up to 20 tokens per second using GPU offloading, Flash Attention, optimized context size, efficient batching, and tuned CPU threading.

Learn how to run GLM-4.7-Flash on an RTX 3090 for fast local inference and integrating with OpenCode to build a fully local automated AI coding agent.

Learn how to fine-tune FunctionGemma in under 10 minutes using Kaggle’s free GPUs, from dataset preparation and baseline evaluation to training and post-fine-tuning validation.

Learn how hook-based automation works and get started using Claude Code hooks to automate coding tasks like testing, formatting, and receiving notifications.

Learn what Claude Cowork is and how to use for file organization, document generation, and browser automation. Hands-on tutorial with real examples and limitations.

A hands-on, end-to-end guide to fine-tuning T5Gemma-2 (270M-270M) for LaTeX OCR, showing how to correctly train and run inference with a multimodal encoder–decoder model using a small dataset.

A simple step-by-step guide to running FLUX.2 locally for high-quality image generation and editing.

Explore what’s new in Claude Code 2.1 by running a set of focused experiments on an existing project repository within CLI and web workflows.

Learn how to build a Gradio Poster Studio App with Qwen-Image-2512 that generates e-commerce creatives with strong typography, clean layout, and photorealistic product scenes.

Learn how to run the SAM Audio base model locally and experience state-of-the-art audio segmentation by isolating voices and sounds with simple, intuitive prompts on an RTX 3090 GPU.

Fine-tune Qwen3-VL-8B on electronic schematics using LoRA. Learn the full pipeline from data preparation to publishing your model on the Hugging Face Hub.

Learn to extract textual data from documents and images using Mistral's latest and most capable OCR model, Mistral OCR 3.