AI Intelligence Archive

You change the AI. Or you open a fresh session. The files are still there. The repository is...

We wrote a writing rule for our agent in July. It read well, it was correct, and for sixty-eight...

llama.cpp vs Ollama: Ollama wraps llama.cpp for convenience, raw llama.cpp wins on speed and control. Benchmarks, GPU offload flags, and when each wins.

The best local LLM for coding by VRAM bracket: Qwen3 Coder vs DeepSeek at 8, 12, 16 and 24 GB, the right quant per card, plus a runnable Ollama setup.

The bottleneck moved, and most teams haven't noticed. It isn't writing anymore. It's proving. I...

Every MCP demo looks the same. Someone spins up a local server, points a client at it, watches a tool...

Your API doesn't need to know who's calling. It needs to know if a human is behind the call. The...

Ok, so, maybe you're a software developer or data scientist who just heard about the new, massive...

The "best self-hosted AI code review tool" question gets searched constantly, and almost everything...

Most chat interfaces have three states: empty, loading, and finished. Agent workflows have many...

I gave my Actor a GitHub MCP connector. It could see 44 tools and use 4. I publish 22...

I connected my audit Actor to Claude, and it audited three packages nobody asked for I...

Explore how diffusion models are revolutionizing AI code generation. Learn the technical architecture, latent space strategies, and engineering challenges of moving beyond autoregression.

Anthropic and Adaptyv Bio have launched a co-sponsored protein design competition that will...

Toda semana alguém me pergunta: "qual IDE com IA você usa?". A resposta curta é "Cursor". A resposta...

Anthropic’s public work on AI R&D automation points to a more demanding way to track advanced AI...

OpenAI has announced Astra for Law, a legal AI foundation that combines its GPT-6 Astra flagship...

Giving an OpenCode Coding Agent Persistent, Editable Memory AI coding agents are good at...

An AI model wrote 69 tests for a Python module. Every one passed. Together they caught zero of the...

Diogo Almeida, who co-invented RLHF and InstructGPT at OpenAI, spent two years building Jev: a System One model that cannot generate text, only typed decisions with calibrated confidence. TypeSafe AI prices input at $0.042/MTok, output fre…

What separates a workflow engine from an orchestrator skill, what that one inversion buys, and how little of what is sold as an engine actually qualifies.

I have a small agent that handles one piece of routine work at a time. It looks at what needs doing,...

A 27B model on my own GPU audited my Splunk and Sysmon install with zero cloud: 10 real findings, and it fixed some of them itself, including two false positives from its own earlier audits.

Debug algorithmic trading backtests with Python: catch look-ahead bias, compare complex models fairly, and test execution costs and order failures.

I started this experiment for a slightly different reason. I was playing with a system that could...

Everybody's shipping AI agents. Nobody's auditing their toolchains. The MCP (Model Context Protocol)...

A model predicts a 92% risk of hospital readmission. The SHAP values show age, prior admissions, and...

Large language models have become remarkably capable. With enough parameters, data, and training,...

Explore why developers are shifting to local-first AI, privacy-focused hardware, and edge computing in 2026 over traditional cloud platforms.

Coding agents have emerged as a promising paradigm for robot manipulation: a language model writes the robot controller as a program, and agents built in this way now operate robots without robot-specific training.Whether this paradigm is …

Embedding spaces define notions of semantic similarity and distance. We study whether those embeddings reflect physical measurements of mass, distance, time and volume, which admit a unique, objective notion of semantic equivalence and dis…

Complex robotic manipulation tasks frequently require a long-term memory of past events and actions. As conditioning on full histories renders policies prone to spurious correlations and degrades performance, many approaches to policy memo…

Modeling articulated objects from sparse monocular views is challenging because each observation reveals only partial geometry and motion evidence. Most feed-forward methods infer articulation from a single observation and therefore rely h…

Professional design requires any-color control: the ability to specify an object's target color with any 24-bit hex value for image generation and editing. Prior work has explored color generation, editing, and colorization, but often reli…

Hereditary polyposis syndromes can be precursor lesions to colorectal cancer and are associated with a broad spectrum of extracolonic tumors. Early identification and accurate classification of these syndromes are essential for timely diag…

Pretraining a neural PDE surrogate can reduce the amount of new CFD data needed when geometry or modeled physics changes. However, it remains unclear how different components of distribution shift affect this benefit. We pretrain a surroga…

Frontier coding agents are increasingly trusted to work autonomously for long periods, yet an agent's final response is often the only account of that work a user sees. We quantify the propensity of frontier agents to \emph{overclaim} task…

Hostile rhetoric toward social groups can normalize exclusion and justify mistreatment, as well as contribute to rising polarization and political violence. Efforts to moderate hostile rhetoric in online speech draw on foundational theorie…

Reinforcement learning (RL) of large language models is notoriously sensitive to small differences between training and inference engines, often referred to as the training-inference mismatch (TIM). However, completely eliminating TIM is i…