Nemotron Lightning 3.5 30B A3B carries NVIDIA's full checkpoint name for the Lightning release of August 11, 2026, where A3B denotes the roughly 3 billion parameters that activate out of a 30 billion parameter Mixture-of-Experts. Interleaved Mamba-2 and MoE layers, with attention used sparingly, are what let it serve at small-model speed. NVIDIA publishes it under OpenMDW-1.1 with the training data and recipes alongside the weights, so it can be modified and deployed commercially without a fee, and it is distributed in BF16 and NVFP4 builds for different accelerator budgets. It is built for the high-frequency, tool-heavy steps of an agent loop, the kind of work where per-call latency matters more than raw capability.
Nemotron Lightning 3.5 30B A3B runs on 1 route through the qynio gateway. Compare residency, zero retention, and training posture at a glance — full data-handling detail per route below.
| Provider | Region | Zero data retention | Training | Input | Output |
|---|---|---|---|---|---|
| Multi | Zero data retention | No | $0.05 | $0.20 |
Nemotron Lightning 3.5 30B A3B runs on a single route through the qynio gateway today, so its availability is that provider's availability.
Name a second model in the same request and the gateway tries it on retriable errors, so a busy hour never has to reach your users. .
Measured over the last 30 days from each provider's official status feed via StatusGator. Refreshed hourly. See uptime for every provider qynio monitors.
Each route hosting Nemotron Lightning 3.5 30B A3B has its own privacy posture, residency, and no-KYC terms. Postures are maintained by qynio with a last-verification timestamp.
Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. GLOBAL; wallets; deposit available.
Call Nemotron Lightning 3.5 30B A3B through the qynio gateway with one API key. Let your coding agent set it up, or call it directly — qynio is drop-in compatible with the OpenAI, Anthropic, and Google AI SDKs.
Copy this and paste it into a coding agent like Claude Code, Cursor or Codex and it'll wire up qynio for you.
import OpenAI from "openai";const client = new OpenAI({apiKey: process.env.QYN_API_KEY,baseURL: "https://api.qyn.io/v3/compat",});const completion = await client.chat.completions.create({model: "fireworks/nemotron-lightning-3p5-30b-a3b",messages: [{ role: "user", content: "Hello" }],});console.log(completion.choices[0].message.content);
Side-by-side on privacy, crypto rails, pricing, and benchmarks.
