Nemotron 3 Super
by NVIDIA · nvidia/nemotron-3-super-120b-a12b
NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...
- Input / 1M tokens
- $0.30
- Output / 1M tokens
- $2.40
- Context
- 1.05M
Specs
| Input price | $0.30 / 1M tokens |
|---|---|
| Output price | $2.40 / 1M tokens |
| Cached input | $0.15 / 1M tokens |
| Context window | 1.05M tokens |
| Max output | 262K tokens |
| Reads | text |
| Writes | text |
| Tool calling | Yes |
| Reasoning | Yes |
| Listed | Mar 11, 2026 |
List prices in USD, updated Oct 30, 2026. Providers may charge differently.
Providers
3 providers serve this model, as of Oct 30, 2026, 18:12 UTC.
| Provider | Quantization | Context | Max output | Input / 1M | Output / 1M | Cache read / 1M | Uptime (30 min) | Latency (median) | Throughput (median) | Supports | Status |
|---|---|---|---|---|---|---|---|---|---|---|---|
| DeepInfra | bf16 | 262K | 16.4K | $0.085 | $0.40 | — | 100% | 5.77 s | 63 tok/s | Normal | |
| DekaLLM | fp8 | 262K | 236K | $0.08 | $0.45 | — | 100% | 517 ms | 48 tok/s | Normal | |
| Crusoe | fp8 | 1.05M | 262K | $0.30 | $2.40 | $0.15 | 100% | 308 ms | 102 tok/s | Normal |
"Cheapest" is the lowest input plus output price; "Fastest" the highest median throughput. Columns empty for every provider are not shown.
Related models
- NVIDIA
Nemotron 3.5 Super VL
- Input
- $0.15
- Output
- $0.50
- Context
- 1.05M
- NVIDIA
Nemotron 3.5 Super VL (free)
- Input
- Free
- Output
- Free
- Context
- 1M
- NVIDIA
Switchyard
- Input
- Varies
- Output
- Varies
- Context
- 1M
- NVIDIA
Nemotron 3.5 Lightning
- Input
- $0.07
- Output
- $0.20
- Context
- 262K
- NVIDIA
Nemotron 3.5 Lightning (free)
- Input
- Free
- Output
- Free
- Context
- 1M
- NVIDIA
Nemotron 3.5 Content Safety
- Input
- $0.20
- Output
- $0.20
- Context
- 131K