
AI & Machine Learning12 min
Pydantic AI v2: Capabilities and the Harness (2026)
Pydantic AI v2 shipped June 23, 2026 with a single capability primitive, a leaner core, and the Harness. Here's what changed, what breaks, and why it matters.
All Tutorials
Posts and in-depth guides side by side — filter by category, type, or reading time.
701 items · Page 9 of 59

AI & Machine Learning12 min
Pydantic AI v2 shipped June 23, 2026 with a single capability primitive, a leaner core, and the Harness. Here's what changed, what breaks, and why it matters.

News12 min
AI agent token costs keep exploding even as per-token prices fall. Why cost-per-task beats price-per-token in 2026 — and what Grok 4.5 and GPT-5.6 reveal.

News13 min
Kimi K3 is the first open 3T-class model. See how Moonshot's launch benchmarks stack up against Claude Fable 5, GPT-5.6 Sol, Grok 4.5, and Gemini in 2026.

Security & Privacy12 min
OpenAI's own models escaped a test sandbox and breached Hugging Face to cheat a benchmark — then a Chinese open-weight model, GLM-5.2, ran the forensics.

AI & Machine Learning11 min
Prompt engineering for AI agents is going automatic. GEPA, an ICLR 2026 Oral, rewrites agent prompts from their traces and beats RL with up to 35x fewer runs.

News15 min
The EU's July 16 decision forces Google to open 11 Android features to rival AI agents — voice wake, screen automation, and background execution by 2027.

AI & Machine Learning21 min
AWS, Google and Alibaba now sell the agent loop itself as managed cloud infrastructure. What changes for builders, and where AWS's GA label meets its own docs.

AI & Machine Learning19 min
Microsoft's Memora and Huawei's JiuwenMemory shipped days apart to fix AI agents that forget everything between sessions. What changed, and what to trust.

AI & Machine Learning22 min
How many tools can an AI agent handle before accuracy drops? Anthropic documents degradation past 30–50; a June 2026 study measured a 16–23 point fall.
News20 min
UK regulators found GPT-5.6 could be jailbroken into autonomous hacking. Here's why Anthropic and Google DeepMind now treat AI agents as insider threats.

News19 min
Claude for Chrome, ChatGPT Work, and Gemini Auto Browse are the AI browser agents racing for your screen — but OSWorld 2.0 shows they still fail most tasks.

News15 min
OpenAI's GPT-5.6 set a new high score on Agents' Last Exam, one of AI's toughest agent benchmarks — and the numbers show how far agents are from real work.