NNVIDIA · Released Jun 4, 2026
Nemotron 3 Ultra
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...
- Released
- Jun 4 2026
- Context window
- 262K 183K out
- Input / 1M tokens
- $0.60 from Free
- Output / 1M tokens
- $2.40 from Free
Where to run it
26 providers compared
Free to try on Nvidia, Kilo Gateway, UnoRouter, Requesty, AIHubMix +2 more — free endpoints are often limited-time and rate-limited. Cheapest: routing.run at $0.10 / $0.10 per 1M input / output tokens — 97% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| Nvidiafree · may be limited-time | Free | Free | — | 1M | 66K | 100%24h | Free |
| Kilo Gatewayfree · may be limited-time | Free | Free | — | 1M | 66K | — | Free |
| UnoRouterfree · may be limited-time | Free | Free | — | 1M | 128K | — | Free |
| Requestyfree · may be limited-time | Free | Free | — | 1M | 66K | — | Free |
| AIHubMixfree · may be limited-time | Free | Free | — | 1M | 128K | — | Free |
| Bothubfree · may be limited-time | Free | Free | — | 1M | 128K | — | Free |
| OpenCodefree · may be limited-time | Free | Free | Free | 1M | 128K | — | Free |
| routing.run | $0.10 | $0.10 | — | 131K | 32K | — | Cheapest |
| DeepInfraFP4 | $0.40 | $1.76 | $0.08 | 262K | 16K | 100%24h | +640% |
| Kilo Gateway | $0.40 | $1.76 | $0.08 | 203K | 183K | — | +640% |
| Ollama Cloud | $0.10 | $3.00 | $0.10 | 262K | 128K | — | +725% |
| CoreWeave | $0.50 | $2.15 | $0.10 | 262K | 262K | — | +813% |
| DeepInfraFP4 | $0.50 | $2.20 | $0.10 | 262K | 16K | 100%24h | +825% |
| DevPass (LLM Gateway) | $0.50 | $2.20 | $0.10 | 1M | 128K | — | +825% |
| NanoGPT | $0.50 | $2.50 | $0.25 | 1M | 66K | — | +900% |
| Pioneer | $0.50 | $2.50 | $0.15 | 1M | 65K | — | +900% |
| Requestyno tool calling | $0.50 | $2.50 | — | 262K | 131K | — | +900% |
| BaseTenus · FP4 | $0.60 | $2.40 | $0.12 | 203K | 183K | 99.3%24h | +950% |
| BaseTenFP4 | $0.60 | $2.40 | $0.12 | 203K | 183K | 98.9%24h | +950% |
| Vercel AI Gateway | $0.60 | $2.40 | $0.12 | 1M | 65K | — | +950% |
| Fireworks AI | $0.60 | $2.40 | $0.12 | 262K | 128K | — | +950% |
| CommandCode | $0.60 | $2.40 | $0.12 | 1M | 128K | — | +950% |
| DigitalOcean | $0.90 | $1.70 | $0.18 | 131K | 131K | — | +1000% |
| VeniceFP8 | $0.63 | $3.13 | $0.19 | 256K | 33K | 100%24h | +1150% |
| Venice AI | $0.63 | $3.13 | $0.19 | 256K | 33K | — | +1150% |
| Together AI | $0.60 | $3.60 | $0.20 | 512K | 512K | — | +1250% |
| Nebius Token Factory | $1.00 | $3.00 | $1.00 | 1M | 1M | — | +1400% |
| Eden AI | $1.00 | $3.00 | $1.00 | 1M | 128K | — | +1400% |
| Kenari | $0.69 | $4.29 | $0.069 | 1M | 128K | — | +1486% |
| Thinking Machines | $1.80 | $6.22 | $0.36 | 66K | 66K | — | +2806% |
Benchmarks
Benchmark scores
No benchmark scores yet — Epoch AI hasn't evaluated Nemotron 3 Ultra. Scores appear here automatically once it does.
Specs
Details
- Model ID
nvidia/nemotron-3-ultra-550b-a55b- Developer
- NVIDIA
- Release date
- Jun 4, 2026 ·
- Context window
- 262,144 tokens
- Max output
- 182,520 tokens
- API price
- $0.60 / $2.40 per 1M input / output tokens
- Providers
- 26
- Input types
- text
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- Yes
- Free tier
- Yes
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.