NVIDIA 把 30B MoE 的 Lightning 放到 Perplexity API 上了，每百万输入才 1 分多钱，适合跑高频工具调用，跟 Ultra 搭配干活很划算。
NVIDIA 的 Nemotron 3.5 Lightning 现已面向所有开发者开放，可通过 Perplexity Agent API 调用。该模型为开源 30B MoE 架构，仅激活 3B 参数，专为高频智能体执行层设计，适用于工具调用、验证和子智能体任务。输入定价为每百万 tokens 0.0115 美元，输出定价为每百万 tokens 0.17 美元。开发者可将 Lightning 与 Nemotron Ultra 等前沿模型搭配，用于规划与批量执行。
Nemotron 3.5 Lightning available to all developers on the Perplexity Agent API. Input: $0.0115 pe...
Nemotron 3.5 Lightning available to all developers on the Perplexity Agent API. Input: $0.0115 per 1M tokens Output: $0.17 per 1M tokens Perplexity Developers @perplexitydevs . @NVIDIA Nemotron 3.5 Lightning is now available in the Perplexity Agent API 🎉 An open 30B MoE model with 3B active parameters, built for the high-volume execution layer of always-on agents: tool calls, validation, and subagent work. Pair Nemotron Lightning 3.5 with frontier models like Nemotron Ultra for planning, and let Lightning handle the volume. Start building: bit.ly/perplexity_lig… 0 🔗 View Quoted Tweet 💬 5 🔄 2 ❤️ 44 👀 7835 📊 7 ⚡