AI

Deep, developer-first coverage of artificial intelligence — from frontier model releases and benchmarks to agents, RAG pipelines, and the AI-native tools changing how we ship software. No hype, just what actually matters to engineers.

Article 2d ago 7

Your Slack Bot's Database Is Now a Claude Session

Vercel's template wires Claude Managed Agents into Chat SDK, and the thin bridge is the point.

Rachel Goldstein

An 8B Fine-Tune Now Fits in 4 GB of VRAM

Soup streams frozen decoder layers through the GPU one at a time, and proves the arithmetic is bit-exact.

Article · 3w ago1

DeepSeek V4 Flash on One AMD GPU Took Nine Patches

A single 192 GB MI300X now serves the full checkpoint — proof of AMD's hardware edge and its lingering software gap.

Article · 3w ago1

Your SSD Is the New VRAM for Local LLMs

Swiftlet streams sparse Qwen experts off disk, fitting an 80B model into 4.3 GB of Mac RAM.

Article · 3w ago1

Your SSD Is the New VRAM

Swiftlet streams MoE experts from SSD, fitting an 80B Qwen in 4.3GB of RAM — and a 35B on an iPhone.

Article · 3w ago1

OpenAI Just Priced Mathematical Discovery at $200 a Problem

Astra's Lean-verified proofs are real, but the token bill and a rival model's one-day replication tell the bigger story.

Article · 3w ago0

Qwen's 16-Day Coding Run Sets a New Bar for Receipts

The public git trace matters more than the model — and the orchestration harness is the part worth stealing.

Article · 3w ago0

$Cloudflare's Quantization Math Is Right. The Disclosure Isn't$

Cloudflare's Quantization Math Is Right. The Disclosure Isn't

FP8 caches and INT4 weights are becoming table stakes; what providers tell you at purchase time is the real fight.

Article · 3w ago0

Quantize the Decode, Not the Prefill

Cloudflare's production numbers for Kimi K2.6 and GLM 5.2 show where trillion-parameter inference actually gets cheaper.

Article · 3w ago5

Rust Finally Moves to Kill Pin

A newly accepted 2026 goal adds Move, Forget, and Destruct traits — and quietly fixes async Rust's original sin.

Article · 3w ago5

Why Your COBOL Migration Should Keep the Bugs

American Express researchers show where LLMs actually belong in mainframe modernization: writing the tests, not the Java.

Article · 3w ago4

AirLLM's 4GB 70B Trick Is Real, and Beside the Point

Layer streaming makes giant models fit; disk bandwidth decides whether you'd ever want to run them that way.

Article · 3w ago1

MiniMax H3 Drains the Last Moat in Open Video

Native stereo audio and 2K clips land in ComfyUI day-0 — but the license carves out the US and EU.

Article · 3w ago2

Retyping AI Code Is a Symptom, Not the Cure

A viral plea to hand-copy every line of LLM output gets the mechanism wrong and the diagnosis right.

Article · 3w ago0

Claude Has Spent Two Weeks Rewriting Its Own App

Boris Cherny's runaway Swift-rewrite experiment shows agent quality is now bounded by verifier quality, not model quality.

Article · 3w ago1

GitHub Models Got 29 Days. Plan Your AI Stack Accordingly

Microsoft retired its free inference API with under a month's hard notice, paying customers included. That's the real lesson.

Article · 3w ago0