Skip to main content
TokenCost logoTokenCost

Nemotron 3 Ultra 550B

Last verified: June 7, 2026

$0.500/1M input · $2.50/1M output · 1.0M context · NVIDIA

Count Tokens

—
Tokens
—
Input Cost
—
Output Cost

Estimate Monthly Cost

Monthly Cost Estimator

Quick:
< $0.0001/mo

Pick a preset above or enter custom usage

Alternatives to Nemotron 3 Ultra 550B

Cheaper Alternative
GLM-4.7
$0.600 in · $2.20 out · Zhipu
7% cheaper
Premium Upgrade
Kimi K2 Thinking
$0.600 in · $2.50 out · Moonshot
Premium tier from Moonshot
Same Provider
Nemotron 3 Super 120B
$0.300 in · $0.800 out · NVIDIA
63% cheaper in NVIDIA ecosystem

Pricing Details

NVIDIALargest open NVIDIA model. Hybrid Mamba-Transformer with latent MoE, 550B total / 55B active, NVFP4 pre-training. No single rate (open weights): $0.50/$2.50 on OpenRouter and DeepInfra, as low as $0.37/$1.08 at launch, $0.60/$3.60 on Together. Free rate-limited endpoint on OpenRouter. AA Intelligence Index 48, top US open-weights model but behind Kimi K2.6 (54). MMLU-Pro 86.8, GPQA 87.0, LiveCodeBench v6 89.0, SWE-Bench Verified 71.9, RULER@1M 94.7. OpenMDW 1.1 license.
Input / 1M tokens
$0.5
Output / 1M tokens
$2.5
Context Window
1.0M
Max Output
33K
Full Pricing TableToken Calculator

Price History

Launched at current price on 2026-06-04. No price changes recorded.

Frequently Asked Questions

Common questions about Nemotron 3 Ultra 550B pricing and usage

Read More About Nemotron 3 Ultra 550B

Nemotron 3 Ultra has no list price. The same open weights run from $0.37 to $0.60 a million depending on who hosts them.
June 7, 2026 · 8 min read
Full Pricing TableCompare ModelsCost Calculator