llm
parsehawk/parsehawk
Local-first document AI. Run 100% locally by default, with API, CLI, and Web UI.
repoholasoymalva/guia-de-programacion-con-ai
Guia de desarrollo y programacion con herramientas de Inteligencia Artificial y LLMS
repoMoh4696/free-ai-models
10 legit ways to access major AI models (GPT, Claude, Llama, DeepSeek & more) for FREE — step-by-step guides, no subscription, no card.
repoteixeirazeus/fablize-for-hermes
This project adapts fablize's verified procedures — verification grounding, multi-story evidence gating, investigation protocol, and early-stop prevention.
repoNoshkoto/Noshy
Your agent has amnesia. Noshy fixes that. Persistent memory for AI agents — LLM extraction, hybrid search, zero deps.
repolikithamadala/ai-pm-learning-fleet
A self-running fleet of AI agents that teaches its owner to think like an AI product manager - and maintains itself daily with no babysitting. Built on Claude Code.
repolas7/memharness
Bi-temporal agent long-term memory: SQLite-backed MCP server with recall ranking, hybrid vector+FTS recall, and a source-staleness signal
repoengineering87/llm-atlas
Interactive, in-browser visualization of how a transformer language model works: tokens, attention, quantization, and sampling, rendered live.
reponomad505050/ions-genesis
Open protocol for traversable knowledge networks. CBB traversal lets an 8B model match frontier AI on domain-specific queries.
reponicolbl95/MultiAgent-Financial-Analyzer
Multi-agent AI system that analyzes financial reports, identifies risks, and generates executive summaries using LangGraph, RAG and LLMs.
glossary_termContext window
How much text a model can consider at once.
companyNorthwind AI
An applied-research lab building reliable reasoning models.
repominimal-diffusion-lm
A tiny, readable diffusion language model.
articleWhy small models are having a moment
Distillation and better data are closing the gap with giants — here's what it means.
paperSparse attention at million-token context
A linear-cost attention variant that holds quality past a million tokens.
paperDiffusion language models, revisited
Bringing diffusion objectives to text generation with competitive quality.
modelHelix-7B
A 7B general model with strong math and code benchmarks.
modelRetrace-1.5B
A small reasoning model tuned to self-correct via failure traces.
modelCodeForge-15B
A code model trained on permissively licensed repositories.
paperScaling laws for mixture-of-experts models
How sparse expert routing changes the compute-optimal frontier for large models.
glossary_termTransformer
The neural network architecture behind most modern language models.
tutorialFine-tune a small model on your own data
A practical guide to LoRA fine-tuning end to end.
tutorialBuild your first transformer from scratch
Step through attention, MLPs, and training on a toy task.
glossary_termToken
A chunk of text (often a word-piece) that a model reads and predicts.
toolPromptLab
An editor for testing and versioning prompts.
glossary_termAttention
A mechanism that lets a model weigh which parts of the input matter most for each output.
paperGrokking in small transformers
When and why tiny models suddenly generalize long after overfitting.
repoattention-zoo
Implementations of many attention variants, benchmarked.