← Back to model catalog

NVIDIA (USA)

Nemotron 3 Ultra

COpen

NVIDIA's October 2026 open-weight MoE model (550B total / 55B active). It drew 6.38T tokens on OpenRouter in a week (#7) and 22.5T over a month (+31%), and a free endpoint with a 1M context makes it a cost-effective pick for long-context batch workloads.

API Pricing

Input

$0.5

Output

$2.2

Cached input

—

Peak hours

—

USD / 1M tokens

※ OpenRouter measured (2026-10): input $0.50 / output $2.20 (USD per million tokens); the :free endpoint of the same name is free (context 1M)

Price sourceopenrouter.ai ↗build.nvidia.com ↗

Latest change:Up 208.8% vs previous($0.4 → $1.235 / 1M)

Price history

12 snapshots · 2026-04-28 → 2026-08-17

Current: $1.235Period low: $0.4 (2026-04-28)Period high: $1.235 (2026-08-17)First tracked: $0.4 (2026-04-28)

Specifications

Context window
262K Tokens
Total parameters
550B
Active parameters
55B
Architecture
MoE
License
NVIDIA Open Model License

Overall score

Total

42.8

Rank

#25

Tier

C

Data confidence

25%

Capability35
Scale35
Value35
Context34.5
Ecosystem55
Freshness94.4

Capabilities & use cases

Best for

  • Private self-hosted deployment
  • Long-context batch processing
  • Enterprise intranet inference
  • High-concurrency inference services

Official links

Similar price

FAQ

What does Nemotron 3 Ultra cost?
Nemotron 3 Ultra is priced at $2.2 output and $0.5 input (USD per 1M tokens)
How large is the context window of Nemotron 3 Ultra?
Nemotron 3 Ultra has a 262K Tokens context window with up to — output.
Where can I access Nemotron 3 Ultra?
Official channels: Website and API platform.

Data is refreshed daily and cross-checked against two independent sources; please confirm pricing and specs on the vendor page.