Nemotron 3.5 Lightning
Nemotron 3.5 Lightning is a large language model made by NVIDIA, added to the catalog on 11 Aug 2026. It has a context window of 1,048,576 tokens and costs $0.07 per 1M input tokens and $0.20 per 1M output tokens. It accepts text and produces text. Typical uses: processing long text inputs, reasoned text responses and tool-assisted agentic workflows.
Price checked ยท source: OpenRouter
What Nemotron 3.5 Lightning is
NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA. It accepts text and produces text. The provider describes it as suited to high-throughput agentic workloads and specialized tasks. Its listed configuration has 3 billion active parameters out of 30 billion total. The model supports reasoning, tool use, structured outputs, and response formats.
The model has a context length of 1,048,576 tokens and a maximum output of 235,929 tokens. This gives it room to process long text inputs while allowing lengthy generated responses, within those stated limits. Its text-only input and output mean that the listed interface is for text rather than image or audio content.
These capabilities suit workflows that pass written instructions or source material to a model, request a reasoned response, and need a defined output format. Tool support is relevant to agentic workflows that incorporate tools; structured outputs and response-format support can be used where an application expects a particular structure. The facts do not specify individual tools, particular schemas, programming languages, or task-level performance, so suitability depends on the surrounding workflow and its requirements. For long-context text work, its stated context and output limits are characteristics to assess alongside expected input and response formats.
Key features
- Accepts text and produces text.
- Context length of 1,048,576 tokens.
- Maximum output of 235,929 tokens.
- Supports reasoning, tools, structured outputs, and response formats.
- Open mixture-of-experts model with 3B active parameters out of 30B total.
Typical uses
- Processing long text inputs
- Reasoned text responses
- Tool-assisted agentic workflows
- Generating structured outputs
Nemotron 3.5 Lightning pricing
US dollars, list price through OpenRouter.
| Item | Standard | Free |
|---|---|---|
| Input, per 1M tokens | $0.07 | $0 |
| Output, per 1M tokens | $0.20 | $0 |
| Cached input read, per 1M tokens | $0.04 | n/a |
| Variant | Model id |
|---|---|
| Free | nvidia/nemotron-3.5-lightning:free |
A variant is the same model under another id, with prices or limits of its own. Its prices are the columns of the list above.
No price change since it was added on 11 Aug 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0001 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0019 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0025 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.011 |
Nemotron 3.5 Lightning specifications
| Model id | nvidia/nemotron-3.5-lightning |
|---|---|
| Version | nemotron-3.5-lightning-20260807 |
| Context window | 1,048,576 tokensAbout 786,432 English words |
| Max output | 235,929 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 11 Aug 2026 |
| Input types | text |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | No |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Nemotron 3.5 Lightning yet.
Compare Nemotron 3.5 Lightning
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.