Nemotron 3 Nano 30B A3B
NVIDIA released 2025-12-14
$0.087 per million tokens, balanced
DeepSeek V4 Flash 0423 is both better and cheaper.
It scores +27.6 higher and costs 30% less ($0.061/M against $0.087/M) on this workload — and it does everything this model does.
Solar Pro 4 is cheaper still (40% less) but drops 262K → 131K max output.
Every price dimension
| Input | $0.050/M |
|---|---|
| Output | $0.200/M |
| Cached input | $0.030/M |
Independent scores
| Intelligence | 14.5 |
|---|---|
| Coding | 14.4 |
| Agentic | 2 |
Sources are listed separately rather than averaged. Across the 62 models both have scored they correlate at r = 0.806 — close agreement overall, but four models rank very differently between them, and a blended score would hide exactly those.
Capability
| Context window | 262K |
|---|---|
| Max output | 262K |
| Input modes | text |
| Tool use | yes |
| Reasoning | optional |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-08-24 |
| Quality data | verified |
NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...