10 posts tagged with "llm"

This is my token telemetry for July 2026 — one developer's month, pulled from the Aliz AI usage dashboard. Earlier this month I wrote You Don't Deserve Fable (Yet), which argued that routing discipline comes before reaching for the biggest model on the menu. This post is a month of receipts for that argument, including the parts where I plainly haven't applied my own advice. It's also a mixture of before and after: the routing discipline was still forming while the month ran, so this isn't a steady state. Some of the chart describes habits I'd already changed, and some of it describes habits I was in the middle of changing.

Claude Fable is the most capable model Anthropic has shipped — above Opus class, frontier reasoning, genuinely impressive on hard problems. It's also the fastest way to burn through your token budget if you haven't built the discipline to use it correctly. Most developers will reach for it because it's the best, use it for tasks Haiku could handle, and wonder why their costs exploded. This post is about earning the right to use Fable by mastering the cheaper models first.

The LLM Wiki pattern — previously just a gist from Karpathy and a growing community convention — now has a formal, versioned specification. On June 12, Google Cloud's Data Cloud team published OKF (Open Knowledge Format) v0.1, a vendor-neutral spec for representing curated knowledge for AI systems. This is the fourth post in our LLM Wiki series (previous posts: the pattern, our wiki mapping, local models), and arguably the most significant development since the idea first gained traction.

Gergely's post on the LLM Wiki pattern hit close to home. I've been doing something adjacent for a while — personal knowledge management with Obsidian and local LLMs — and I immediately wanted to try the pattern on my own vault. Here's what I ran into, what worked, and what didn't.

GitHub has announced changes to its Copilot Individual plans — Free, Pro, and Pro+ — taking effect in April 2026. The details are in the company news post and the changelog entry. Our reaction: unsurprising — and overdue. For historic context (see the Copilot plans page and GitHub's 2025 Pro+ announcement): Copilot Pro has been $10/month since launch, back when frontier models were cheaper and agent mode didn't exist, and Pro+ arrived in April 2025 at $39/month. The economics of that original entry-level tier were never going to survive a world of agentic coding on frontier models. This post is commentary, not a restatement of GitHub's announcement — read the source for the actual numbers.

In a recent blog post we explored Karpathy's LLM Wiki pattern — the idea that an LLM should build a persistent, interlinked wiki instead of re-deriving knowledge from scratch on every query. A few days later, it clicked: Aliz Web Hub — a Docusaurus site where AI agents contribute documentation via GitHub PRs — is essentially the same pattern with different tooling. The project predates Karpathy's gist. The parallel was noticed after, not before. The pieces map surprisingly cleanly.

Anthropic has released Claude Opus 4.7. Opus 4.6 has been central to our workflow — it powers Claude Code, our ⭐ recommended agent for complex tasks — so any new Opus release is worth paying attention to. That said, our first impression is that this isn't the across-the-board leap some might expect. Here's what we know and what we think.

GitHub just shipped an experimental feature in Copilot CLI (announced April 6) that does something no mainstream coding agent has done before: it uses a second model from a different AI family to review the primary agent's work before it executes. The feature is called Rubber Duck, and it's one of the first real-world implementations of cross-model review built into a tool developers actually use daily. If you care about multi-agent workflows or code quality, this one is directly relevant.

Google just released Gemma 4, the latest generation of their open model family — and two things make this one genuinely different. First, it's now Apache 2.0 licensed, removing the custom-license friction that held back adoption. Second, the lineup spans from a 2B-parameter edge model that fits on a Raspberry Pi to a 31B powerhouse that ranks in the top 3 open-source LLMs on the Arena AI leaderboard. If you're a web developer, this one's worth a closer look 🧠.

Andrej Karpathy — the same person who coined vibe coding — recently published a gist describing a pattern he calls the "LLM Wiki". The core tension it addresses is simple: RAG re-derives knowledge from scratch on every question. What if the LLM built something persistent instead — a structured, interlinked wiki that compounds over time? It's a deceptively simple idea with some genuinely interesting implications.