Nemotron 3 Ultra 550B A55B
DeprecatedNemotron 3 Ultra 550B A55B is NVIDIA's language model with a 1.0M context window and up to 461K output tokens, available from 8 providers, starting at $0.25 / 1M input and $2.15 / 1M output. A 550B-parameter mixture-of-experts Nemotron model with 55B active parameters, built for frontier-scale reasoning, tool use, and agentic tasks.
| Benchmarks | |
|---|---|
| Intelligence Index | #136 |
| Coding Index | #73 |
| GPQA | #79 |
| HLE | #99 |
| IFBench | #4 |
| Time to First Token | #414 |
| SciCode | #165 |
| LCR | #56 |
| TerminalBench Hard | #58 |
| TAU2 | #106 |
| Output TPS | #67 |
Capabilities
Input2/5
Text✓
Image✓
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities7/13
Reasoning✓
Adaptive Reasoning·
Function Calling✓
Parallel Function Calling✓
Structured Outputs✓
Native JSON Schema✓
Web Search✓
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching✓
Assistant Prefill·
Pricing by Provider
US Dollar ($)
Per 1M tokens
| Provider | Standard | Free | |||
|---|---|---|---|---|---|
| Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | Input $ / 1M | Output $ / 1M | |
| $0.6 | $2.40 | $0.12 | — | — | |
| $0.5 | $2.20 | $0.1 | — | — | |
| $1.00 | $3.00 | N/A | — | — | |
| $0.6 | $2.40 | $0.12 | N/A | N/A | |
| — | — | — | $N/A | $N/A | |
| $0.25 | $2.50 | $0.25 | — | — | |
| $0.6 | $3.60 | $0.2 | — | — | |
| $0.6 | $2.40 | $0.12 | — | — | |
| $0.5 | $2.15 | $0.1 | — | — | |
Cost Calculator
US Dollar ($)
Preset:
Cheapest Instances to Run It
Cloud GPU instances that can host Nemotron 3 Ultra 550B A55B, ranked by cheapest on-demand price. The model needs about 1320 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_ND96is_MI300X_v5 |  Azure | 8× AMD Instinct MI300X GPU (192GB) | 1536 GB | $48.00/hr | ||
| Standard_ND96isr_MI300X_v5 | Azure | 8× AMD Instinct MI300X | 1536 GB | $48.00/hr | ||
| p6-b200.48xlarge |  AWS | 8× B200 | 1432 GB | $113.93/hr | ||
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| Nemotron 3 Ultra 550B A55B | 1.0M | $0.250 | $2.15 | Current | |
| Nemotron 3 Ultra NVfp4 | — | 262K | $0.660 | $2.64 | Available |
Other Models
| Model | Tier | Released | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| Nemotron 4 15B | — | — | — | — | — |
| Nemotron Lightning 3.5 | — | 1.0M | $0.050 | $0.200 | |
| Nemotron 3.5 Lightning 30B | — | 1.0M | — | — | |
| Nemotron 3.5 Content Safety | — | 131K | $0.200 | $0.200 | |
| Nemotron Nano 3 30B A3B Omni Reasoning | — | 256K | — | — | |
| Nemotron Super 3 120B | — | 256K | $0.150 | $0.650 | |
| Nemotron Nano 3 30B | — | 262K | $0.060 | $0.240 | |
| Nemotron Nano 3 30B A3B Reasoning | — | — | — | — | — |
| Nemotron Nano 3 30B A3B Omni | — | — | — | — | — |
| Nemotron Nano 3 4B | — | — | — | — | — |