Nemotron Super 3 120B A12B
Nemotron Super 3 120B A12B is NVIDIA's language model with a 1.0M context window and up to 262K output tokens, available from 2 providers, starting at $0.15 / 1M input and $0.455 / 1M output. A 120B-parameter hybrid MoE Nemotron Super 3 model with 12B active parameters, optimized by NVIDIA for compute-efficient reasoning in specialized agentic systems.
Capabilities
Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities5/13
Reasoning✓
Adaptive Reasoning·
Function Calling✓
Parallel Function Calling·
Structured Outputs✓
Native JSON Schema✓
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching✓
Assistant Prefill·
Pricing by Provider
US Dollar ($)
Per 1M tokens
| Provider | Standard | ||
|---|---|---|---|
| Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | |
| $0.21 | $0.455 | $0.06 | |
| $0.15 | $0.65 | N/A | |
Cost Calculator
US Dollar ($)
Preset:
Cheapest Instances to Run It
Cloud GPU instances that can host Nemotron Super 3 120B A12B, ranked by cheapest on-demand price. The model needs about 288 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.
All clouds
FP16 (full precision)
US Dollar ($)
Instance | Cloud | GPU | VRAM | Price | Cheapest region | |
|---|---|---|---|---|---|---|
| Standard_NC96ads_A100_v4 |  Azure | 4× NVIDIA A100 | 320 GB | $14.69/hr | westus2 | |
| g7e.24xlarge |  AWS | 4× RTX PRO Server 6000 | 384 GB | $16.57/hr | us-east-1 | |
| p4d.24xlarge | AWS | 8× A100 | 320 GB | $21.96/hr | us-west-2 | |
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| Nemotron Super 3 120B A12B | 1.0M | $0.150 | $0.455 | Current | |
| Nemotron Nano 3 30B A3B | 262K | $0.050 | $0.200 | Available | |
| Nemotron 3 Ultra | — | — | $0.600 | $2.40 | Available |
| Nemotron Nano 2 12B | 131K | $0.200 | $0.200 | Available | |
| Nemotron Nano 2 9B | 131K | $0.040 | $0.160 | Deprecated |