NVIDIA Nemotron 3.5 Lightning Weights Now on Hugging Face
MaziyarPanahi · x · 2026-08-11
NVIDIA's Nemotron 3.5 Lightning model weights are now available on Hugging Face. The model is a 30B MoE with 3B active parameters, designed for high-throughput tasks, delivering up to 4x the output speed of similar-sized models.
Related event: NVIDIA Open-Sources Nemotron Model and Agent Router(37 posts)→
More from Models
- NVIDIA Reportedly Building 1-Trillion-Parameter Nemotron 4 Model — scaling01 · 2026-08-12
- Where Do Quantized Local LLMs Break? Reddit Users Share Experiences — d77chong · 2026-08-12
- Unreleased Anthropic Model Makes Surprising Progress on the Riemann Hypothesis — TechCrunch AI · 2026-08-12
- Frontier Models' Biggest Bottleneck is 'Lack of Self-Esteem' in Math Research — coherence · 2026-08-12
- Pokee-Isaac 28B Beats Meta and Qwen Rivals in Sub-30B Agent Benchmarks — Kyrannio · 2026-08-12
- Post-Trained NVIDIA Nemotron Beats Claude Opus in Legal Agent Tasks — ctnzr · 2026-08-12