- likes
- 42
- comments
- 3
Post
NVIDIA Nemotron 3.5 Lightning is live on Baseten day 0! Nemotron 3.5 Lightning is the fastest open model in its class, built for long-running agents. - 4x higher throughput compared to similar-sized open models - 50% lower-cost and 63.4% fewer output tokens in production testing - Wins the accuracy-versus-speed Pareto frontier on the AA Intelligence Index - 30B MoE architecture (3B active) - 1M token context length CodeRabbit fine-tuned it using Baseten Training and achieved 80.4% routing agreement, a 4.6% increase over the baseline 75.8%.