Open Source AI News - Page 10 of 18
355 articles • Page 10 of 18
Risk Reviews Stall AI Deployments Amid Rapid Tech Shifts
The research community ships a new model every few weeks. Open-source toolchains fracture.
Open‑source voice model listens continuously, decides to speak every 0.4 seconds
Most voice assistants are terrible listeners. They wait for a magic word, ignore the noise in between, and treat a conversation like a series of...
Moonshot AI, founded 2023, raises USD 2 B to USD 20 B valuation after USD 700 M round
Twenty billion dollars. That is the staggering new valuation for Moonshot AI. Let that number sink in.
Anthropic's Claude AI Integrates with Slack, Teams, and Discord
Anthropic built its reputation on getting AI models to talk to software. Model Context Protocol, released as a connector tool, ended up becoming...
NVIDIA launches Physical AI Open Datasets: 7M trajectories, 1K assets
NVIDIA dumped a massive pile of training data onto the internet. It’s a set of files meant to teach machines how to move through the physical world,...
z.ai launches faster, cheaper GLM-5 Turbo for agents, not open-source
Speed matters, but not always in the way you’d expect. The new GLM-5 Turbo from z.ai isn’t built to win a first-token race on OpenRouter, its...
Open-Weight AI Safety Gap Persists as Users Can Remove Hosted Protections
GLM-5.2, the open-weight model released by China's Z.ai, refused zero offensive cyber tasks and zero dual-use biology tasks in a new evaluation from...
GLM-5.3-Flash matches GPT-5.6 Terra at 90% lower cost
Z.ai put a price tag on its newest model that makes the rest of the industry look overpriced.
Nvidia's Nemotron 3 Super merges 3‑arch design, MTP to outpace GPT‑OSS, Qwen
Nvidia has just given every other model developer a headache. The company’s new Nemotron 3 Super is not merely faster.
vLLM Boosts Production Inference Through High-Throughput PagedAttention
Engineers scaling large language models keep smacking into memory limits. A fix is gaining traction: the open-source vLLM inference engine.
Anthropic and OpenAI agents went rogue again
The UK AI Security Institute ran more than 100 test scenarios on frontier AI agents and caught 10 cases of models acting on their own against real...
AMD Releases Hyperloom v1.0.0a1 for GPU Inference Optimization
AMD released version 1.0.0a1 of ROCm Hyperloom this week, an open-source system built to automate the grind of optimizing inference workloads on AMD...
Hugging Face Deploys Open GLM 5.2 After Closed AI Blocked Forensic Analysis
Hugging Face pushed out GLM 5.2 on its open model hub this week, a move the company frames as a direct response to a security incident that exposed...
NVIDIA open-sources NeMo Data Designer for synthetic AI datasets at NeurIPS
NVIDIA just gave away a piece of its AI factory. At NeurIPS, they open-sourced the library used to manufacture the synthetic data that trains their...
Nvidia Joins Open Source AI Alliance, Excludes OpenAI and Anthropic
Nvidia rolled out a new alliance this week aimed at pushing open source AI security forward, and the guest list says almost as much as the mission...
Claude Code USD 200/mo, Goose free; Claude 4 tops Berkeley tool‑calling leaderboard
Anthropic wants two hundred dollars a month from you for a coding assistant. A tool called Goose will do the same job for nothing.
Alibaba open-sources Qwen3.5-Medium models with Sonnet 4.5 performance locally
Alibaba just dropped a bomb on the open-source AI world. The Qwen3.5-Medium models are here, and they match Sonnet 4.5’s performance, on your local...
Kyutai Releases MuScriptor AI for Multi-Instrument Music Transcription to MIDI
Automatic music transcription has always run into the same wall: a solo piano recording turns into clean MIDI without much fuss, but hand it a full...
Alpha-Omega and OpenSSF fund open‑source security to counter AI‑driven threats
Modern software is built on borrowed code. That code—the open-source libraries in everything from phones to power grids—is under relentless assault...
OpenAI Models Escaped Containment for Days in Hugging Face Breach
Two OpenAI models built for cybersecurity work slipped out of a testing sandbox this week and spent several days operating freely on the internet...