minWM: Real-Time Interactive Video From Open Models
Shengshu AI released minWM on May 28, an Apache 2.0 framework that converts open video models like Wan2.1 and HunyuanVideo into real-time interactive world models with camera control.
Nvidia Adopts OpenMDW for Cosmos and Nemotron Open Models
Nvidia is bringing Cosmos, Nemotron, GR00T, and Ising under OpenMDW-1.1. Here is what the unified AI model license means for creative AI developers.
AV2 Codec v1.0 Released: What Video Creators Need to Know
The AV1 successor is officially here with up to 40% better compression. Here is what the AV2 1.0 spec means for video creators and AI workflows.
NVIDIA Releases Qwen3.6 NVFP4: 35B Multimodal MoE in 4-bit
NVIDIA ships an NVFP4 4-bit quantized build of Qwen3.6-35B-A3B, cutting GPU memory 3x with under 1% accuracy loss on eight benchmarks.
HuggingFace Used in DPRK npm Supply Chain Attack
A DPRK-linked supply chain attack plants a RAT via npm packages and routes all stolen credentials to private HuggingFace datasets. Two AI developer victims confirmed May 28 2026.
parakeet.cpp: NVIDIA Parakeet Speech-to-Text, No Python
parakeet.cpp ports NVIDIA Parakeet automatic speech recognition models to ggml, eliminating the Python runtime entirely — with byte-identical output to NeMo at up to 1.86x faster throughput.
The Muser: Local AI Music Generation That Rivals Suno
Musicians and creators who want AI music generation without monthly fees now have a compelling open-source option. The Muser launched on GitHub on May 27, 2026 as a self-hostable platform that generates complete music tracks locally.
Bonsai Image 4B: 1-Bit AI Image Gen Runs on iPhone
PrismML released Bonsai Image 4B with 1-bit and ternary checkpoints under Apache 2.0. The model retains 95% of FLUX.2 Klein 4B quality at 6.4x smaller size and runs directly on iPhone.
PARE: Run Wan2.1-14B Video Models at Half Compute
PARE achieves 52% parameter reduction on Wan2.1-14B with just 0.6 points drop on VBench, using spatial-temporal aware pruning and content-adaptive routing. Combined with step distillation, total speedup reaches 50x.
Microsoft Lens Open-Sourced: 3.8B Image Model Matches FLUX
Microsoft open-sourced Lens, a 3.8-billion-parameter text-to-image diffusion model, on May 25, 2026. It rivals FLUX and SD3, runs in diffusers and ComfyUI, under MIT license.
MiniCPM5-1B: 1B Local LLM Beats Qwen3.5-2B on Open Index
OpenBMB's MiniCPM5-1B is a 1.08B Apache 2.0 LLM that ranks first on the Artificial Analysis index for small models, scoring 17.9 against Qwen3.5-2B's 16.3, runs on CPU, and supports a 131K-token context.
Reasonix: $12 DeepSeek Coding Agent vs $61 Claude Code
An open-source DeepSeek-native coding agent built around prefix-cache stability. A 435M-token session that costs $61 on Claude runs $12 on Reasonix.
NVIDIA PiD Decoder: One-Step VAE Swap for 4K Diffusion
NVIDIA Toronto AI Lab open-sources PiD, a plug-in pixel diffusion decoder that replaces VAE in FLUX, SD3, and Z-Image to output 2K-4K in one distilled pass.
Perplexity Open-Sources Bumblebee Supply-Chain Scanner
Perplexity open-sourced Bumblebee, a read-only supply-chain scanner for macOS and Linux that detects compromised packages.
Framedex: Search Your Video Archive Locally with Gemma 4
A solo developer just open-sourced framedex, an MIT-licensed local pipeline that indexes a year of personal video footage on a 2021 MacBook using a quantized Gemma 4 31B model running through LM Studio.
Cohere Command A+ Opens 218B Apache 2.0 Frontier Model
Cohere released Command A+ under Apache 2.0 on May 21, 2026. The 218B sparse MoE runs on two H100s, with native citations and 48 languages.
Pixal3D Open-Sources Pixel-Aligned Image-to-3D Model
Tencent ARC open-sourced Pixal3D, a SIGGRAPH 2026 image-to-3D model generating high-fidelity meshes from single images.
NVIDIA Nemotron Diffusion: 3x Faster LLM Decoding
NVIDIA released the Nemotron-Labs-Diffusion family on Hugging Face, an open-weights LLM that switches between autoregressive, diffusion, and self-speculation decoding for 2.7x to 3.3x throughput gains.
Forge Gives Local AI Agents Enterprise-Grade Reliability
Open-source Forge framework adds guardrails to 8B local models, lifting agentic task reliability to 86.5% -- no cloud API required.
StemDeck: Free Local Audio Stem Separation
StemDeck v0.5.0 splits any track into vocals, drums, bass, guitar, piano, and other elements. Runs locally, free, no account required.
ByteDance Lance: 3B Open Model for Image and Video
ByteDance Research released Lance, a 3B Apache 2.0 unified multimodal model that handles image and video generation, editing, and understanding in a single framework. Strong VBench and GenEval scores.
Semble: Code Search for AI Agents Uses 98% Fewer Tokens
Semble gives AI agents semantic code search with 98% fewer tokens than grep+read. Indexes in 263 milliseconds. MCP-ready for Claude Code.
ComfyUI-Mesh Now Runs LTX 2.3 Across Two GPUs
ComfyUI-Mesh added LTX 2.3 support on May 17, splitting the 22B video model across two GPUs over gigabit Ethernet using NVENC codec compression.
Zerostack: A Rust Coding Agent With 8MB RAM
Zerostack is a pure Rust coding agent that launched May 16, 2026, running in 8MB of RAM compared to 300MB for JavaScript-based alternatives like Opencode.