Model overview
NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...
The summary below is compiled from sources by an AI model, not our own testing. Verify selection decisions against original sources and your workload.
Key features
- Open mixture-of-experts (MoE) architecture with 3B active parameters per forward pass
- Optimized for high-throughput agentic workloads and specialized reasoning tasks
- Free-to-use model available via OpenRouter
Limitations & considerations
- Total parameter count is 30B, but only 3B are active per inference — may limit expressiveness on some complex tasks
- Limited public documentation on fine-tuning support, tool integration details, or multimodal capabilities
- No information provided about context window handling beyond the stated 1M context size claim
Source prices & cost analysis
Prices vary by channel, region and configuration. Channel quotes are listed independently and are not treated as a single official provider price.
| Source / channel | Input | Output | Billing unit | Collected at |
|---|---|---|---|---|
| OpenRouter · nvidia/nemotron-3.5-lightning:free ↗ | USD 0 | USD 0 | Per million tokens | 2026/11/2 06:25:22 |
Estimate monthly base token costs
Estimates exclude cache discounts, tiered pricing, multimodal surcharges, web search, taxes and additional reasoning output.
Public benchmarks & run configuration
暂无已采集的可核验公开评分。模型未上榜不代表能力较低。
View full indexed rankings →Frequently asked questions
Is Nemotron 3.5 Lightning truly free to use?
Yes — according to the source, it is offered as a free model on OpenRouter.
What does '3B active parameters out of 30B total' mean?
It is a sparse mixture-of-experts model: only 3 billion parameters are activated per token generation, though the full model contains 30 billion parameters.
Does it support tool use or function calling?
The model is tagged with 'tools', suggesting designed support for tool-augmented workflows, but specific implementation details (e.g., JSON schema adherence, built-in tool registry) are not disclosed in the provided material.
Do open weights permit free commercial use?
Not necessarily. Check the version license, commercial conditions and usage policies. Self-hosting incurs compute and maintenance costs.
Are these real-time billing prices?
These are timestamped source snapshots, not live billing prices. Check the channel for current versions, regions, discounts and billing methods.
Sources & traceability
许可证：来源未披露，须进一步核验。模型介绍与第三方评测均以原来源的具体版本和方法为准。