Chunk Tuner is a tool that benchmarks chunking strategies for a RAG corpus and recommends a configuration. It benchmarks strategies such as fixed-token windows, recursive character splitting, semantic splitting and PDF structural chunking, scoring each against retrieval metrics. It is available as a Python library, a CLI and an MCP server.
A blog post describing a small library for building agents controlled by large language models, inspired by langchain. The agent is instructed by a prompt and runs in a loop of Thought, Action and Observation. Implemented tools include executing Python code in a REPL, Google search and Hacker News search.
An index page from LabLab for discovering AI technologies and related topics. Each technology page is described as offering tutorials, upcoming hackathons, frameworks and boilerplates. Listed technologies include GPT-4, ChatGPT, BERT, AI21 Labs, Anthropic, AWS, Cohere, Hugging Face and DeepSeek.
gotoHuman is a platform for supervising AI agents, letting workflows request human review of critical steps. Reviewers approve and edit AI-generated content in an Agent Inbox, with email and Slack notifications, and completed reviews return to the workflow via webhook. Review types are configured in a no-code editor and can capture text, images, markdown and JSON.
Accord.NET is a .NET machine learning framework combined with audio and image processing libraries written in C#. It covers classification, regression, clustering, distributions, hypothesis tests and kernel methods, with imaging and signal processing components. The framework is intended for building computer vision, audition, signal processing and statistics applications, including commercial use. It offers sample applications, documentation and a wiki.
A visualisation of major large-language models ranked by performance using the MMLU benchmark, which consists of 16,000 multiple-choice questions across 57 academic subjects. It also presents selected lawsuits filed against AI companies and notes that MMLU has some critiques.
An Ahead of AI article by Sebastian Raschka sharing lessons from hundreds of experiments on finetuning LLMs with low-rank adaptation (LoRA), billed as an essential technique for efficiently training custom open-source LLMs. It covers LoRA consistency, QLoRA compute-memory trade-offs, learning rate schedulers, Adam versus SGD, multiple epochs, enabling LoRA for more layers, and balancing rank and alpha.
Community notes
Be the first to share how this project worked for you.
Sign in with a verified account to leave a review or join the discussion.
Sign in to reviewKeep exploring
A blog post describing a small library for building agents controlled by large language models, inspired by langchain. The agent is instructed by a prompt and runs in a loop of Thought, Action and Observation. Implemented tools include executing Python code in a REPL, Google search and Hacker News search.
An index page from LabLab for discovering AI technologies and related topics. Each technology page is described as offering tutorials, upcoming hackathons, frameworks and boilerplates. Listed technologies include GPT-4, ChatGPT, BERT, AI21 Labs, Anthropic, AWS, Cohere, Hugging Face and DeepSeek.
gotoHuman is a platform for supervising AI agents, letting workflows request human review of critical steps. Reviewers approve and edit AI-generated content in an Agent Inbox, with email and Slack notifications, and completed reviews return to the workflow via webhook. Review types are configured in a no-code editor and can capture text, images, markdown and JSON.
Accord.NET is a .NET machine learning framework combined with audio and image processing libraries written in C#. It covers classification, regression, clustering, distributions, hypothesis tests and kernel methods, with imaging and signal processing components. The framework is intended for building computer vision, audition, signal processing and statistics applications, including commercial use. It offers sample applications, documentation and a wiki.
A visualisation of major large-language models ranked by performance using the MMLU benchmark, which consists of 16,000 multiple-choice questions across 57 academic subjects. It also presents selected lawsuits filed against AI companies and notes that MMLU has some critiques.
An Ahead of AI article by Sebastian Raschka sharing lessons from hundreds of experiments on finetuning LLMs with low-rank adaptation (LoRA), billed as an essential technique for efficiently training custom open-source LLMs. It covers LoRA consistency, QLoRA compute-memory trade-offs, learning rate schedulers, Adam versus SGD, multiple epochs, enabling LoRA for more layers, and balancing rank and alpha.
