#open-weights

Meta released Muse Glimmer, a 30B open-weights model designed to run AI agents on consumer hardware, signaling Zuckerberg's vision of distributed superintelligence.

NVIDIA's 364M-parameter multilingual text-to-speech model gains Arabic, Korean, and Brazilian Portuguese support, enabling low-latency voice agents on private infrastructure.

A 2.6B-parameter model trained for tool use and multi-step reasoning, matching performance of models 4x larger while running at 220 tokens/sec on consumer hardware.

Moonshot AI's release of Kimi K3 weights for free signals a shift in how Chinese firms are challenging US dominance in large language models.

US tech leaders split sharply on whether to restrict Chinese open-weight models, with 200+ startups opposing a ban while safety-focused giants push for controls.

Major AI firms urge policymakers to distinguish legitimate model-development techniques from IP theft

A U.S. open-source AI lab challenges the narrative that Chinese models like Qwen and Kimi K3 are vectors for state-sponsored compromise, citing technical constraints on backdoor insertion.

Chinese firm Moonshot AI released Kimi K3, sparking Wall Street sell-offs and renewed geopolitical tensions over open-weights models and AI leadership.

The startup founded by former OpenAI executives releases its flagship multimodal model, challenging the dominance of closed-source systems.

Chinese AI lab Moonshot is preparing to release Kimi K3, an open-weights model expected to match closed-source frontier models, as investor confidence in open alternatives grows.

Inkling combines native image, audio, and text processing with a 1M-token context window and sparse MoE architecture for efficient multimodal reasoning.

NVIDIA releases three open-weights embedding models, with the 8B variant claiming the #1 spot on RTEB's multilingual leaderboard and targeting enterprise agentic retrieval workloads.

Zhipu AI's open-weights model rivals Anthropic's Mythos on vulnerability detection, raising national security concerns amid hardware restrictions.

Peak XV-backed startup releases culturally-tuned video generation model at ₹0.48/second, positioning open-weights AI for mass adoption across India.

DiffusionGemma uses parallel text diffusion instead of sequential token generation, achieving 1000+ tokens/sec on H100 GPUs with trade-offs in output quality.

Google DeepMind releases Gemma 4 12B, a 12-billion-parameter model with unified vision and audio processing that runs on 16GB consumer hardware.

Hugging Face hackathon project demonstrates how tiny language models can power real-time simulations that frontier models cannot economically support.

JetBrains' new Mixture-of-Experts model achieves 2x speedup over dense peers while activating just 2.5B parameters per token.

Liquid AI unveils a sparse 8-billion-parameter model with 1-billion active parameters, trained on 38T tokens—a scale comparable to frontier model training runs.

NVIDIA releases diffusion language models at 3B, 8B, and 14B scales that generate and refine tokens in parallel, offering latency improvements for GPU-constrained inference workloads.

Stability AI releases four new audio models capable of generating full-length songs, with open-weights tiers and licensing deals backing the release.

Allen Institute releases OlmoEarth v1.1, a more efficient earth-observation model family that maintains v1 performance while reducing compute through shorter token sequences.