nvidia Open weights Reasoning Free tier

Nemotron 3 Ultra (free) pricing & benchmarks

Released Jun 4, 2026 · served by 1 provider · ranked #37 by intelligence in our index

Input / 1M tokensFree
Output / 1M tokensFree
Context window1M
Cached input
Intelligence index37.8
Coding index49.3
Agentic index27.4
1M in + 300K out$0.00

What Nemotron 3 Ultra (free) is

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 and prices for Nemotron 3 Ultra (free)

The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.

ProviderInputOutputContext QuantUptime 24hThroughput
Nvidiacheapest Free Free 1M 99.6%

Nemotron 3 Ultra (free) benchmark results

Design Arena head-to-head results, as reported through the OpenRouter model API.

CategoryArenaRankEloWin rate
svg models #49 1128 37.6%
3d models #50 1190 41.8%
asciiart models #50 1112 37.5%
uicomponent models #59 1180 39.5%
gamedev models #60 1183 38.8%
codecategories models #69 1155 36.2%
dataviz models #71 1151 37.4%
website models #82 1127 32.5%

FAQ

How much does the Nemotron 3 Ultra (free) API cost?

Free per 1M input tokens and Free per 1M output tokens.

Is Nemotron 3 Ultra (free) free?

Yes — it is served at $0 for both input and output tokens on OpenRouter, subject to rate limits.

Which provider is cheapest for Nemotron 3 Ultra (free)?

Nvidia at Free input / Free output per 1M tokens.

Can I self-host Nemotron 3 Ultra (free)?

Yes — weights are published as nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 on Hugging Face, so you can run it on your own hardware.