Open-Source AI
Open models — Llama, Mistral, DeepSeek, Qwen and others — are catching up to closed labs and let you run AI on your own hardware, vendor-free. That shifts the balance of power, the safety debate and the economics of the field. Here is everything on open-source AI: open-weight releases, licenses, local deployment and the race against closed models.
The End of Openness: Meta Replaces Llama With the Closed Muse Spark Model
Meta stunned the AI community by dropping Llama in favor of the proprietary Muse Spark. The move to a closed model puts thousands of startups at risk and changes…
LLM as Decoder in ASR: Adapting Nvidia's SALM Architecture for Russian-Language Telephony
Kontur adapted Nvidia's hybrid SALM architecture for Russian-language telephony and video conferencing. Instead of a classic cascade system, they used an LLM (Qwen)…
Deep Fusion Memory cuts retrieval 32x and beats baselines by 12.7%
Deep Fusion Memory injects external knowledge into a frozen LLM's hidden states with no retrieval at inference: 32x fewer queries and up to +12.71% on QA tasks.
Synchronous MoE decoding makes GPUs wait for the busiest group
MTEdit: 500 thousand editing trajectories teach models not to lose quality over 8 steps
Conditional Program shrinks DeepSeek's n-gram memory from 203 to 51 GB at 2 bits
RoboHarness: A Coding Agent on Qwen3.8-Flash-Next Beats VLA Models in BEHAVIOR
Same 74 Tasks, Different People: Browser AI Agents Lose 36% of Their Success Rate
Researchers from nine universities gave browser AI agents identical tasks but had different simulated "people" phrase the requests. The average score fell from…
AIPOCH Open-Sources Open-Science, a Lab Notebook for "Vibe Research" with AI
AIPOCH has released Open-Science as open source. The project promises to make AI-assisted research traceable, transferable and reproducible. Implementation details…
NVIDIA RTX Spark N1X Runs a Local Qwen3.8 27B Agent in 18.7 Minutes
Vera Rubin NVL72 beats GB300 by up to 5.7x in tokens per megawatt in AA-AgentPerf
NVIDIA releases Nemotron 3.5 Super: 26 AA points and 630 tokens/s
Apple's LensVLM compresses long agent context into images and teaches the model to "zoom"
Google Uses Gemini 4 Argon to Refactor Chromium and Its Millions of Lines of Code
Google engineer Eric Chen runs several Gemini 4 Argon agents to untangle Chromium's architecture: millions of lines of code, with test suites and human review as…
Gemini 4 Argon Teaches Apollo 2 Humanoid to Play Cornhole: 7 Hits in a Row
Google DeepMind showed Gemini 4 Argon building a MuJoCo simulation from a video, after which the Apollo 2 humanoid hit the target seven times in a row. Weeks of…