Fable 5 Returns as Claude Sonnet 5 Becomes Default
Fable 5 is back with tighter safeguards while Sonnet 5 becomes Claude's default. Here is the practical routing and cost decision.
Posts
Articles
Technical notes, experiments, and field reports on AI agents, coding tools, security failures, and verification workflows from real production use.
How to Read AI Benchmarks Without Getting Fooled
Learn how to read AI benchmarks, spot saturation and contamination, compare coding and chat leaderboards, and test models on your own work.
GPT-5.6 Sol Preview: Access and Agent Risks
GPT-5.6 Sol is a restricted API and Codex preview shaped by a government request. Check access, agent risks, pricing, and safeguards before routing work.
Frontier Model Access: Fable, Mythos, GPT-5.6
Fable 5 is restored globally, while Mythos 5 and GPT-5.6 remain approval-sensitive. These are different access mechanisms, not one licensing regime.
Codex Banked Resets and the GPT-5.6 Preview
Codex usage resets can accumulate for manual use. Check /usage before waiting or buying credits; GPT-5.6 remains a limited preview.
Claude Fable 5 Restored: Cost and Guardrails
Fable 5 returned globally July 1. Compare its $10/$50 API price, temporary plan access, safeguards, retention, and practical alternatives.
AI Coding Subscription Limits: May 2026
May 2026 coding-limit snapshot. Current discovery routes Kimi K3 as the newest flagship and K2.7 Code as the cheaper coding API.
Running LLMs Locally: Gemma, Ollama, and Practical Choices
How to decide between local and cloud, where Gemma fits, which runtime to pick, and what hardware you actually need for useful local AI.
Exa Search: The AI-Native Search Engine Built for Agents
Deep dive into Exa, the AI-native search API. Real benchmarks: 452ms latency, 79-90% token savings, semantic discovery vs keyword matching.
Claude Code Leak: The Hidden Features That Escaped, and What Builders Should Actually Care About
The Claude Code leak exposed hidden features, but the deeper lesson is operational: package artifacts can leak product boundaries, internal experiments, and roadmap intent.
AI Tools in March 2026
March 2026 AI tool recommendations by budget tier — free options, $20/mo, under $100, and $200+. Updated for the current free tier landscape.
Migrating from OpenAI to Kimi, Claude, or Antigravity (2026)
Complete migration guide: export your OpenAI data, compare features with Kimi k2.5 and Claude, cut costs by 80%.
OpenClaw Provider Policy Check 2026: Complete Guide
Find out if your AI provider allows OpenClaw usage. 10-second status check for Anthropic, Google, OpenAI, Kimi, and 10+ providers with migration paths if you're at risk.
Hands-on with Google Lyria 3: AI Music Generation in the Gemini App
First look at Google Lyria 3 — accessible right inside the Gemini app. Includes the revised Lyria prompt workflow, an ancient Mesopotamian trap experiment, image-to-music, and a cat theme song.
Anthropic Blacklisted by Pentagon: What Happened
Anthropic was formally designated a 'supply chain risk' and blacklisted from all U.S. government work on February 27 — hours before OpenAI announced a Pentagon deal containing the same safety redlines Anthropic was punished for holding.
Google Whisk first look: subject-scene-style video generation (experimental)
Hands-on with Google Whisk: a three-ingredient approach to AI video generation — subject, scene, style — with 3 free generations and surprising results when you guide it well.
OpenCode GLM-5 Free Tier Snapshot
Historical snapshot of OpenCode's GLM-5 free-tier moment, updated to point readers toward GLM-5.1 and avoid stale leaderboard claims.
Google Flow: the best free AI text-to-video tool I've found in 2026
Hands-on with Google Flow and Veo 3.1 Fast: generating professional video ads and animated GIFs from images and text prompts — 50 free credits per day, no subscription needed.
Anthropic OAuth Policy Feb 2026: What Changed
Anthropic's official Claude Code compliance docs explicitly prohibit OAuth tokens in third-party tools, including the Agent SDK. Here's what's new, why OpenCode broke, and where the OSS ecosystem goes next.
UltraThink → UltraQuiet: Why Devs Want Receipts
Claude Code started hiding file paths and search details behind Ctrl+O. Here's a working fix: the AGENTS.md Audit Ledger pattern.