Nemotron 3 Ultra (free)
LLM
by nvidia ·
nvidia-nemotron-3-ultra-550b-a55b-20260604
Context window
1M
Max output
66K
Providers
4
Best input
Free
About this model
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...
Providers & prices
| Provider | Input $/MTok | Output $/MTok | Context | Upstream ID |
|---|---|---|---|---|
| N NVIDIA | Free | Free | 1M | nvidia/nemotron-3-ultra-550b-a55b:free |
| D DeepInfra | $0.5 | $2.20 | 262K | nvidia/nemotron-3-ultra-550b-a55b |
| B BaseTen | $0.6 | $2.40 | 203K | nvidia/nemotron-3-ultra-550b-a55b |
| V Venice | $0.625 | $3.12 | 256K | nvidia/nemotron-3-ultra-550b-a55b |