# NVIDIA models — API pricing and providers

> 4 NVIDIA models with NVIDIA's list rates (checked 2026-10-05). Page: https://infro.io/models/nvidia

| Model | Id | Modality | List rate | Providers |
| --- | --- | --- | --- | --- |
| [Nemotron 3.5 Lightning 30B A3B](https://infro.io/models/nvidia/nemotron-3.5-lightning-30b-a3b) | `nvidia/nemotron-3.5-lightning-30b-a3b` | Text | $0.05 per 1M input tokens and $0.20 per 1M output tokens, with cached input at $0.01 per 1M tokens | 2 |
| [Nemotron 3 Ultra](https://infro.io/models/nvidia/nemotron-3-ultra) | `nvidia/nemotron-3-ultra` | Text | $0.60 per 1M input tokens and $2.40 per 1M output tokens, with cached input at $0.12 per 1M tokens | 2 |
| [Parakeet TDT 0.6B v3](https://infro.io/models/nvidia/parakeet-tdt-0.6b-v3) | `nvidia/parakeet-tdt-0.6b-v3` | Audio | $0.09 per audio hour | 2 |
| [Nemotron 3.5 ASR](https://infro.io/models/nvidia/nemotron-3.5-asr) | `nvidia/nemotron-3.5-asr` | Audio | $0.27 per audio hour | 2 |

## Questions

### Which NVIDIA models does this reference list?

4 NVIDIA models: 2 text and 2 audio. Each has its own page with NVIDIA's list rate and the providers that serve it.

### Which providers serve NVIDIA models?

Fireworks AI, OpenRouter and Together AI. You connect your own account with any of them; INFRO routes between the ones you connect.

### What is the lowest NVIDIA input rate listed?

Nemotron 3.5 Lightning 30B A3B, at $0.05 per 1M input tokens and $0.20 per 1M output tokens, with cached input at $0.01 per 1M tokens on NVIDIA's list (checked 2026-10-05).

### Does INFRO add a markup to NVIDIA model usage?

No. Your NVIDIA or provider account bills usage directly; INFRO charges a software subscription and takes 0% of model spend.