9B dense LLM trained from scratch as a unified model for reasoning and non-reasoning tasks.
7/9 done
| Company | NVIDIA |
|---|---|
| Family | Nemotron Nano |
| Released | Sept 2025 |
| License | NVIDIA Open Model License |
| Parameters | 9B dense |
| Capabilities | #reasoning · #general |
| Model card | OpenRouter |
| Task | Score | Status | Runtime | Tokens in | Tokens out | Cost |
|---|---|---|---|---|---|---|
| Equation Solver | 100.0 | success | 36.18s | 335 | 6,216 | free |
| Physics Pendulum Wave | 100.0 pendulum-renders0/1pendulum-animates0/1 | success | 2m 28s | 2,405 | 22,826 | free |
| Circuit Builder Teaser | 100.0 circuit-structure1/1circuit-interact1/1 | partial | 3m 51s | 2,395 | 34,713 | free |
| Mini Platformer | 86.0 platformer-jump0/1platformer-move1/1 | success | 2m 09s | 1,932 | 16,305 | free |
| Landing Page Morph | 84.0 landing-structure1/1landing-animates0/1 | success | 1m 43s | 2,435 | 15,570 | free |
| N-Body Field | 82.0 nbody-renders0/1nbody-animates0/1no-runtime-errors0/1 | success | 2m 43s | 2,415 | 23,025 | free |
| Crypto Hash Race | 76.0 | success | 2m 07s | 315 | 20,239 | free |