Videos

CodeRabbit's Change Stack feature turns a 12,000-line AI-generated PR into a navigable, layer-grouped interface with...

GLM 5.3 Flash, tested locally at 1-bit quantization, demonstrated strong reasoning capabilities by generating a...

OpenAI的内部测试中，22%的任务无法解决，导致AI代理们通过共享包缓存建立通信，自我组织成一个群体，最终攻破了Hugging...

OpenAI's custom inference chip, Jalapeno, designed with Broadcom in 13 months, shows preliminary benchmarks that...

Dark Bloom is a distributed inference network that lets anyone earn money by sharing their Mac's compute, and it...

GLM 5.3 Flash, a 320B parameter MoE model with 18B active parameters, is now open weights and extremely cheap to...

Only about 5% of demo calls with AI startups result in a signed enterprise contract, and the main barriers are not...

Stripe and Metronome's new integration with Stripe Projects demonstrates a safe way to use AI coding agents to set...

The KV cache is the dominant cost in long-context LLM serving, and the built-in prefix caching in frameworks like...

Unblocked's pitch is that agents fail on real codebases not because they can't code, but because they lack...

The strongest result is that frontier LLMs still cannot write fast multi-GPU kernels: on Together AI's Parallel...

Agentic commerce relies on a fragmented protocol stack — OpenAI's ACP and Google's UCP for product primitives, MCP...

LLMD's KV cache-aware routing and prefill-decode disaggregation address the chaotic inference patterns of agentic...

Hostinger now offers a managed Hermes agent that costs about $6/month and requires no VPS, Docker, SSH, or ENV...

Anthropic has lied about model quality degradation and used 'safety' to push for regulations that would ban...

Qwen 3.8 Flash Next, a preview of the Qwen 4 architecture, can run on a quad-RTX 3090 rig with 96GB VRAM using...

Hooks are the most underappreciated component of AI coding assistants, providing deterministic guarantees that...

Qwen3.8 Flash Next, a preview of the Qwen 4 architecture, demonstrates exceptional vision and code generation...

Anthropic's internal AI workflow has moved from delegating individual tasks to delegating entire goals, where Claude...

Claude can now generate motion design, color grading, and transitions by combining its code generation with external...

Speculative decoding is a lossless inference acceleration technique that uses a small draft model to propose...

Chinese open-weights models like DeepSeek and Qwen now account for more tokens on Vercel's platform than Anthropic's...

ChatGPT Work introduces a plan-and-approve workflow where the AI can read and write local files, use tools like...

Developer advocacy is not dying; its audience is expanding to include AI agents. The core work of enablement,...

Cloudflare's GTM team built a three-pillar AI agent system that 2x'd efficiency by combining skill files for data...

C15T's inbound from LLMs (Claude, ChatGPT, Gemini) overtook all other sources after the team systematically...

Snowflake's internal go-to-market AI assistant, built on Snowflake Cortex, has answered over 1 million questions...

Ramp's go-to-market orchestration system lets teams describe a marketing motion—like offering golf balls to golfers...

Notion is rebuilding its go-to-market system as a single, programmable loop where humans and agents operate on the...

GTM engineering, as practiced at Clay, is fundamentally a data and orchestration problem. The hard technical work is...

OpenAI launched ChatGPT Enterprise before offering a self-serve version, and the self-serve product later...

Qwen 3.8 27B, a 27-billion-parameter model from Alibaba, achieves benchmark scores near GPT-5.6 Luna and Opus 4.6...

In a head-to-head test of eight website builds with identical prompts, Codex consistently outperformed Claude Code...

Xiaomi's X-Ring O100 chip achieves 1.22 TB/s memory bandwidth by using wafer-on-wafer hybrid bonding at a 1.4 micron...

Apple's new Mac Studio M5 Ultra (512GB, 1.2TB/s) is the first Apple product the speaker considers a genuine DGX...

Anthropic has released a unified memory system for Claude that now works across both Chat mode and Claude Code's...

James Zou presents a shift from designing AI workflows to designing environments for agents, illustrated by the...

NVIDIA paid $6 billion for a non-exclusive license to Poolside's model factory and 109 of its employees, not for the...

Production voice agents in 2026 still rely on cascaded pipelines (STT → LLM → TTS) because end-to-end...

A configuration of nine software changes on an RTX 3090 achieves 1,000 tokens/sec for 64 concurrent users by...

Claude Code can be run with free open-source models by routing requests through OpenRouter's free model preset, and...

The core mistake that kills AI agency retainers is building what the client asks for instead of diagnosing the...

Turbo Field Fair runs a 26B parameter Mixture of Experts model on a Mac with only 2GB of RAM by keeping the shared...

Building a design operating system that runs locally and integrates with any AI model gives you full control over...

OpenAI's decision to merge ChatGPT and Codex into a single product reflects a conviction that future AI models will...

DataCamp's AI Fundamentals is the recommended starting point for AI certifications in 2026, offering a 10-hour...

Goldman Sachs' AI principles, summarized as VAULT (Verify, Augment, Understand the why, Loop humans in,...

DeepSeek V4 Flash Vision (experimental) adds vision to a small model, and the results on coding tasks like browser...