Tag: open source
-
NVIDIA Releases gpt-oss-puzzle-88B: Up to 2.82× Faster Reasoning on a Single H100
On March 26, 2026, NVIDIA released gpt-oss-puzzle-88B — a deployment-optimized version of OpenAI’s gpt-oss-120B reasoning model that delivers up to 2.82× faster inference while matching or exceeding the original model’s accuracy. Created using NVIDIA’s Puzzle neural architecture search framework, the 88-billion-parameter model fits on a single H100 GPU and represents a significant step forward in […]
-
LiteLLM Backdoored, LM Studio Flagged: AI Tools Face Supply Chain Threats
Two major security incidents hit the AI developer ecosystem within hours on March 24, 2026: backdoored versions of LiteLLM, the popular LLM API proxy, were published to PyPI carrying a multi-stage credential stealer, while LM Studio users reported Windows Defender flagging the local AI tool as a trojan. Together, the incidents underscore how the software […]
-
Cursor’s Composer 2 Exposed as Kimi K2.5 Under the Hood
Cursor’s launch of Composer 2 on March 19, 2026 turned into one of the AI industry’s most public attribution scandals when a developer discovered within hours that the “self-developed” coding model was built on top of Moonshot AI’s open-source Kimi K2.5. The incident has reignited debate over transparency, open-source licensing, and what it means to […]
-
Unsloth Studio: Open-Source No-Code UI for Local LLM Training and Inference
On March 17, 2026, Unsloth AI launched Unsloth Studio — an open-source, no-code web interface that lets users train, run, and export large language models from a single local dashboard. The beta release positions Unsloth Studio as the first tool to unify fine-tuning, inference, and deployment in one local application, promising 2x faster training with […]
-
Leanstral: Mistral’s Open-Source Proof Agent for Lean 4
On March 16, 2026, Mistral AI released Leanstral — the first open-source AI agent purpose-built for Lean 4, the formal proof assistant used across mathematical research and verified software development. With 120 billion total parameters but only 6 billion active per token, Leanstral can formally prove that AI-generated code meets its specifications — addressing one […]
-
Mistral Small 4: Four Models Unified in One Open-Source MoE
On March 16, 2026, Mistral AI released Mistral Small 4 — a 119-billion-parameter Mixture-of-Experts model that unifies instruction following, reasoning, multimodal understanding, and agentic coding into a single deployment. With only 6 billion active parameters per token (8B including embedding layers), it delivers frontier-class performance at a fraction of the cost and latency of larger […]
-
NVIDIA Launches Nemotron Coalition to Build Open Frontier AI Models
At GTC 2026 on March 16, NVIDIA announced the Nemotron Coalition — a first-of-its-kind collaboration uniting eight leading AI labs to co-develop open frontier foundation models. The coalition pools expertise, datasets, and compute on NVIDIA DGX Cloud to produce open-weight models that any organization can specialize for its own domain. General Audience Key points At […]
-
NVIDIA Nemotron 3 Super: 120B Hybrid Model Activates Only 12B Parameters for Agentic AI
On March 11, 2026, NVIDIA released Nemotron 3 Super — a 120-billion-parameter open-weight model that activates only 12 billion parameters per token, delivering frontier-class reasoning at a fraction of the compute cost. Designed for multi-agent AI systems, the model combines three distinct architectures into a single hybrid backbone and ships with a native 1-million-token context […]
-
Karpathy Open-Sources Autoresearch: 100 AI Experiments Overnight on One GPU
Andrej Karpathy, former director of AI at Tesla and co-founder of OpenAI, has open-sourced autoresearch — a compact 630-line Python framework that lets AI agents autonomously design, run, and evaluate machine learning experiments on a single GPU. Released on March 7, 2026, the tool has already garnered over 8,000 GitHub stars and sparked a broader […]
-
FlashAttention-4: Algorithm and Kernel Co-Design for Blackwell GPUs
Key points On March 5, 2026, Tri Dao and collaborators from Princeton University, Together AI, Meta, NVIDIA, and Colfax Research released FlashAttention-4, a ground-up redesign of the attention kernel optimized for NVIDIA’s Blackwell GPUs. The kernel reaches up to 1,605 TFLOPs/s on the B200 at 71 percent hardware utilization, running 1.3x faster than cuDNN 9.13 […]