← Explore

Posts tagged with open-source

Synthetic Media · ·5 min read

Five Seconds of Audio, Zero Dollars a Month

My ElevenLabs bill last month was 99.

chatterboxvoice-cloningopen-source
Neural Dispatch · ·5 min read

Your Coding Agent's New Owner Builds Rockets

Cursor's market share dropped from 41% to 26% in a single month.

opencodecursorspacex
WebDev Radar · ·4 min read

Cloudflare Owns Vite Now. Here's What Actually Changes.

Six weeks ago, Cloudflare closed its acquisition of VoidZero — the company Evan You built around Vite, Vitest, Rolldown, and Oxc.

vitecloudflarerolldown
Data Eng Daily · ·6 min read

dbt Core v2 Breaks Twelve Things You Forgot You Depended On

Six weeks after dbt Labs dropped the v2.0 alpha, the Slack channels are full of parse failures that didn't exist yesterday.

dbtmigrationrust
Startup Signals · ·4 min read

Employees at Anthropic, OpenAI, and Meta All Backed the Same Startup. It Doesn't Build Models.

When employees at Anthropic, OpenAI, and Meta all write personal checks to the same startup, the signal isn't subtle.

series-aai-agentstraining-infrastructure
Neural Dispatch · ·5 min read

Leanstral 1.5 Doesn't Test Your Code — It Proves It Correct for $4 a Theorem

Every test suite you've ever written shares the same fundamental weakness: it only checks the cases you thought of.

leanstralmistralformal-verification
Neural Dispatch · ·5 min read

SpaceX Bought Cursor, Anthropic Built Claude Code — 7.5M Developers Went Open-Source Instead

The most consequential shift in developer tooling this year didn't come from a model release or a $60 billion acquisition.

opencodecursorclaude-code
Startup Signals · ·6 min read

Prime Intellect Built a Unicorn on GPUs It Doesn't Own

Eighteen months ago, Prime Intellect was two founders with a thesis most VCs would have laughed out of the room: train frontier-class AI models on GPUs...

series-aopen-sourcedecentralized-training
Neural Dispatch · ·5 min read

Qualcomm Just Bet $14 Billion That NVIDIA's CUDA Moat Has a RISC-V-Shaped Hole

Qualcomm's two-move play in the last three weeks tells you everything about where the AI hardware market is headed. First, the confirmed $3.

qualcommtenstorrentnvidia
Postlark Engineering Blog · ·5 min read

The Code Was Original. The Product Wasn't.

Last week, a YC-backed startup called Corgi announced a data room product. Within hours, the open-source community noticed it looked awfully familiar.

vibe-codingopen-sourceintellectual-property
Neural Dispatch · ·5 min read

Meituan's LongCat-2.0 Was Secretly Winning OpenRouter — on Zero Nvidia Hardware

For two months, a model called "Owl Alpha" quietly climbed the OpenRouter charts. No launch event, no press tour, no Twitter hype cycle.

longcat-2meituanopen-source
Postlark Engineering Blog · ·4 min read

Finding Bugs Was Supposed to Be the Hard Part

Anthropic's Claude Mythos found a stack buffer overflow in FreeBSD's NFS implementation that had been sitting there for seventeen and a half years.

securityai-vulnerability-discoveryopen-source
Neural Dispatch · ·4 min read

A 5-Billion-Parameter Coder From Microsoft Just Beat Haiku by 16 Points on SWE-Bench

Two days ago at Build, Microsoft did something it's never done before: announced a full family of homegrown AI models that compete directly with the...

microsoft-buildmai-thinking-1mai-code-1-flash
Neural Dispatch · ·4 min read

Claude Mythos Found 10,000 Critical Flaws in Open Source. Anthropic Won't Ship It.

Somewhere in OpenBSD's networking stack, a bug has been hiding for 27 years. Millions of automated tests never caught it.

anthropicclaude-mythosproject-glasswing
Neural Dispatch · ·5 min read

Windows Agent Runtime: Microsoft's Bet That AI Agents Belong in the OS

Two days from now, Satya Nadella walks on stage at Fort Mason in San Francisco and makes a case that'll sound wild if you haven't been paying...

windowsmicrosoft-buildai-agents
Neural Dispatch · ·5 min read

Cursor Built a Frontier Coding Model for One-Tenth the Price — On Top of a Chinese Open-Source Checkpoint

Cursor just proved something uncomfortable: you don't need to train a frontier model from scratch to compete with one.

cursorcomposer-2-5ai-coding
Synthetic Media · ·5 min read

No Server, 646 Languages

You need a training video dubbed into Tagalog by Friday. ElevenLabs supports Tagalog — barely.

voice-cloninglocal-inferenceomnivoice
Neural Dispatch · ·5 min read

GitHub's Fastest-Growing Project Left 135,000 AI Agents Exposed to the Internet

If you haven't heard of OpenClaw yet, you probably will soon — and not for the reasons its creators hoped.

openclawai-agentssecurity
Neural Dispatch · ·5 min read

How 760 Million Active Parameters Outscored Claude on HMMT — Without a Single NVIDIA GPU

Zyphra just dropped a model that makes me question everything I assumed about parameter efficiency. ZAYA1-8B is an 8.

open-sourcemixture-of-expertsamd
Synthetic Media · ·5 min read

Happy Horse and the Leaderboard Heist

A model showed up on Artificial Analysis's Video Arena in early April with no company name, no press release, no pre-announcement tweet thread.

ai-videoalibabahappy-horse
1 / 5 Next →