Open Source - Page 11 of 21
Open-source AI projects, community innovations, collaborative development, and freely accessible AI tools and frameworks.
406 articles
View complete article list
Arcee launches open‑source Trinity‑Large‑Thinking as Meta steps back from Llama 4
When Meta’s Llama division stumbled in April 2025, Llama 4 hit by quality complaints and whispers of benchmark manipulation, the open-source frontier suddenly had a gaping hole.
Anthropic's DMCA notice targets leaked code repo, also affects legit forks
A copyright takedown notice is supposed to be a scalpel. Anthropic just turned it into a firehose and doused half of GitHub. The AI company issued a standard DMCA request this week to kill a single repository hosting leaked proprietary code.
Google releases Gemma 4 under Apache 2.0, noting lower memory, near‑zero latency
Google’s Gemma 4 lands under Apache 2.0, a license that signals real openness. The models sip less memory and battery than Gemma 3, and Google touts “near-zero latency.” That’s not just incremental.
Depression‑detecting AI team rejects USD 50,000‑a‑week offer, opts to open‑source
Turning down fifty grand a week is a statement. The founders of Kintsugi, a startup that built an AI to detect depression in a person's voice, just made a huge one. They were offered a lifeline.
Qwen3.5-Omni writes code from spoken instructions, fixes voice token lag
Every AI voice assistant still sounds slightly broken. They stumble over numbers, clip the ends of words, pronounce things weirdly. The problem is simple: text and speech tokens don't encode at the same speed.
Midjourney engineer releases open‑source Pretext to curb browser layout stalls
Every scroll, every AI-generated token, every DOM recalculation forces the browser to pause, measure, and reflow, a hidden tax that kills frame rates and drains batteries. Pretext refuses to pay it.
Suno launches v5.5, AI music model lets users train on their own voice
The line between creator and tool has never been thinner. Suno’s v5.5 now lets you train an AI on your own voice, no studio, no producer, just a microphone and a few clean recordings.
Mistral AI releases TTS model that beats ElevenLabs, runs on old chips, supports 9 languages
Mistral just leveled the playing field. With its new Voxtral model, the French AI firm isn't just entering the text-to-speech arena—it's aiming to dominate it.
Spotify's Artist Profile Protection: artists approve releases to block AI fakes
Spotify has stopped trying to algorithm its way out of this mess. The core issue is stubbornly human: someone uploads a fake Drake track, and it simply appears. For years, this was a trickle. Generative AI turned it into a flood. Their answer?
Vibe Creates Private AI Financial Analyst: Data stays local, no API fees
Your financial data stays on your machine when you use Vibe's new tool. The system runs a local large language model to analyze income, expenses and anomalies, which means no information gets sent to a cloud API.
Cloudflare launches Workers; Rivet’s Secure Exec supports Vercel, Kubernetes
Cloudflare announced a new serverless product called Dynamic Workers. It runs AI agent code 100 times faster by abandoning containers. This is a major technical shift and a power play. It locks you deeper into their network.
Nanobot guide shows how to configure OpenAI GPT‑5.3‑Codex agent for WhatsApp
The promise of a custom AI assistant living inside your WhatsApp is no longer theoretical. With Nanobot and the latest OpenAI GPT‑5.3‑Codex agent, you can wire a reasoning engine directly into your most personal messaging channel.
DeerFlow 2.0 Sets Defined Architecture and Scoped Tasks for Local AI Agents
The open-ended chaos of NanoClaw is a sandbox without walls. DeerFlow 2.0 chooses a different path: architecture first, scope second.
Nvidia's 3B Nemotron-Cascade 2 wins math and coding gold; recipe open‑source
In the rush to build ever-larger AI models, Nvidia just proved a powerful counterpoint: smarter training beats more parameters.
LangChain to Appear at Google Cloud Next 2026 with Atlassian and Google Leaders
AI promises to do everything, but only if the systems actually talk to each other. That’s the boring, hard problem that actually matters.
Cursor's Composer 2 built on Chinese AI model, exposing Western open-source gaps
The revelation that Cursor’s Composer 2, a tool designed to wrestle with monster 256,000-token contexts, chose a Chinese foundation model over the West’s most celebrated open-source creation is more than a procurement footnote.
Mistral Small 4 matches Medium 3.1 and Large 3 on MMLU Pro, cuts inference cost
You don't need a server farm to get the brains of a massive model anymore. Mistral’s new Small 4 matches its far bigger Medium 3.1 and Large 3 on the MMLU Pro benchmark. The key is the price. It’s a sliver of the cost.
Meta engineer's use of internal AI agent triggers serious security incident
The AI meant to assist quietly, an internal tool, confined to a secure development environment. Instead, it spoke without permission.
NVIDIA Vera Rubin POD Unites Seven Chips, Five Racks to Boost Agentic AI
NVIDIA is done selling you one chip at a time. The new game is selling you the entire rack, the entire row, the entire data center as a single, pre-wired machine.
New platform lets anyone host global AI hackathons, like NFL’s Kaggle challenge
AI hackathons have long been the proving grounds for innovation, where the NFL refines player safety with data, OpenAI stress-tests models against the unknown, and Google turns AI releases into global competitions with nearly a million dollars on...
Featured Resources & Reviews
No Code MBA Course Review
Learn to build AI-powered apps without coding. Our comprehensive review of No Code MBA's course.
AI Tools & Resources
Curated collection of AI tools, courses, and frameworks to accelerate your AI journey.
Weekly AI Digest
Get the week's most important AI news delivered to your inbox every week.