NVIDIA models
9 NVIDIA models you can call through UberLLM’s OpenAI-compatible API. The cheapest is NVIDIA Nemotron 3 Nano 30B at $0.0008/1M input tokens, as of October 24, 2026. Every request is routed to the cheapest healthy seller; the price shown includes UberLLM’s 10% fee.
| Model | Price (from) | Context | Type | Capabilities |
|---|---|---|---|---|
| NVIDIA Nemotron Cascade 2 30B nvidia/nemotron-cascade-2-30b-a3b | 256K | Text | ToolsJSON | |
| NVIDIA Nemotron 3 Nano 30B nvidia/nemotron-3-nano-30b-a3b | $0.0008 in · $0.0033 out | 128K | Text | Tools |
| NVIDIA Nemotron 3 Ultra 550B nvidia/nemotron-3-ultra-550b-a55b | $0.0055 in · $0.0275 out | 1000K | Text | ToolsJSON |
| NVIDIA Nemotron Nano 9B v2 nvidia/nemotron-nano-9b-v2 | $0.0012 in · $0.0048 out | 128K | Text | ToolsJSON |
| NVIDIA Nemotron 3 Super 120B nvidia/nemotron-3-super-120b | $0.0009 in · $0.0050 out | 256K | Text | ToolsJSON |
| NVIDIA Nemotron Nano 12B v2 VL BF16 nvidia/nemotron-nano-12b-v2 | $0.0024 in · $0.0073 out | 128K | Text | ToolsVisionJSON |
| NVIDIA Nemotron 3.5 Lightning 30B nvidia/nemotron-3-5-lightning-30b-a3b | $0.0704 in · $0.1760 out | 1000K | Text | ToolsReasoningJSON |
| NVIDIA: Nemotron 3.5 Super VL nvidia/nemotron-3.5-super-vl | $0.0825 in · $0.2750 out | 262K | Text | ToolsReasoningVisionJSON |
| Cosmos Predict (image-to-video) nvidia/cosmos-predict-image-to-video | $0.62 per video | Video |
Compare NVIDIA models
NVIDIA models — FAQ
What is the cheapest NVIDIA model?
NVIDIA Nemotron 3 Nano 30B is the cheapest NVIDIA model on UberLLM right now, at $0.0008/1M input tokens. UberLLM routes each request to the cheapest healthy seller and bills the price shown, which includes UberLLM's 10% fee.
How many NVIDIA models can I use through UberLLM?
9 NVIDIA models are on the UberLLM marketplace today. One OpenAI-compatible API key and base URL (https://api.uberllm.dev/v1) calls every one of them — no per-provider contracts.
How do I call a NVIDIA model with the API?
Use any OpenAI-compatible SDK: set the base URL to https://api.uberllm.dev/v1, your UberLLM key, and the model id (for example "nvidia/nemotron-cascade-2-30b-a3b"). Anthropic SDKs work against https://api.uberllm.dev/v1/messages.