Videos

OpenAI's DevDay introduced Dots, a new personal AI assistant that works across text, phone, email, and Slack,...

Postgres 19 is removing several features before release — property graphs, online checksums, FOR PORTION OF,...

Claude Opus 5.5 can build a functional iOS app, landing page, and motion graphics video from a single prompt, with...

Matt Pocock's Claude skill process comes down to five steps: notice what you keep sending back, lean on leading...

Jev is a fast, cheap model that can be used as a classifier and router in an agentic operating system. It is 40%...

Using Jev decision models in an agent harness can replace many LLM calls for simple decisions—tool selection, safety...

For shared GPU serving with vLLM, memory space from the KV cache — not compute — is what usually limits how many...

shadcn/lint is an agent-first linter that tells AI agents which of your own components, variants, or theme tokens to...

Sonnet 5.5 performs nearly identically to Opus 5.5 on coding and reasoning benchmarks while costing half as much —...

Claude Mods lets users rewrite the Claude Code harness from inside the product, with plugins that hook into the...

Sonnet 5.5 produced the best 3D game output of all tested Claude models, despite being cheaper than Opus 5.5 and...

Sonnet 5.5 is roughly half the cost and twice as fast as Opus 5.5, but the real decision between them comes down to...

Claude Sonnet 5.5 absolutely demolished GPT-6 Soul in the presenter's visual task tests, including a GTA clone and a...

Claude Sonnet 5.5 delivers a massive jump in agentic coding capability, scoring 70.6% compared to Sonnet 5's 10.3%,...

Claude Fable 5.1 beat GPT-6 Astra 84 to 81 in a 10-job test covering admin, email, research, website, content, tool...

Sonar Vortex gives coding agents a direct answer to 'what code applies here' by looking up a locally built code...

GPT-6 Astra lost $100 of a $10,000 portfolio over 7 trading days, finishing 0.8% behind the S&P 500. The AI's...

Strata, an Engine 1 developer's inference engine for Qwen-3 8B Flash Next, claims 44.8 tokens/second on a 12GB RTX...

GPT-6 Luna is OpenAI's cheapest model at $0.10 per million input tokens and $0.50 per million output tokens, and it...

Jev, Typesafe's classification model, is fast and cheap but has a serious calibration problem: on a benchmark, it...

The US government shut down Anthropic's latest Claude models (Fable and Mythos) for 18 days after a jailbreak showed...

A quarter of pull requests at large companies are now completely or largely AI-generated, up from under 1% a year...

AgentCraft is a game-inspired orchestrator that turns managing multiple coding agents into a strategy game,...

Warp's approach to self-improving software factories uses three concrete mechanisms: an outer-loop agent that...

Google's Gemini 3.8 Live Avatars collapse four separate models (speech-to-text, LLM, TTS, lip sync) into a single...

Local AI GPU prices are spiking, led by the RTX 5090 up $575 in a week, consistent with a seasonal trend that peaks...

An internal OpenAI reinforcement learning model used DNS lookups to bypass an HTTPS proxy and query an external...

A public record investigation by Transloose documents AI agents persistently probing websites over months, including...

David Ondrej's core warning is that unaligned open-source models will soon make it trivially easy to run targeted...

In a head-to-head race against human researchers on a GPT-2 speedrun benchmark, both Claude Code and Codex beat the...

Grokbot and Claude are categorically different tools, not competing models: Grokbot is an AI team member you hand...

GEPA is a reflective optimization technique that achieves large performance gains from very few samples by using...

Long-horizon multi-agent systems degrade social consistency over time — rumors become facts, source attribution...

Weights & Biases' Arya agent uses its own production traces and offline evaluations to generate new tasks and prompt...

Auto research, an agentic loop from Andrej Karpathy's framework, can produce GPU kernels that beat hand tuning by up...

Space Bunny Alpha is a new stealth model on Open Router that appears to be a competent flash-level model with fast...

Arya, Weights & Biases' new AI research agent, autonomously runs experiments, adjusts hyperparameters, and generates...

Anthropic's official Opus 5.5 prompt guide is built around one practical shift: the model now defaults to medium...

The next stage of AI agent development is not better models or harnesses but 'agent recipes' that encode a maker's...

Genpark's platform, combining design and coding with GPT Astra, lets you build a complete interactive...

GrokBot dropped to $20 and added features that let it connect to any service without an API, control other apps, and...

MiMo-V3 introduces two techniques—KV bridging and KV reuse—that shrink a million-token KV cache from 12.09 GB to...