ModelPickr

Models / Nvidia / Nemotron 3 Ultra 550B A55B

Nvidia · nemotron

Nemotron 3 Ultra 550B A55B

Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy

$0.50 in · $2.50 out

per 1M tokens · first-party price. Reference price from Nvidia, the lab's own listing.

Provider input prices run from $0.10 to $1.00 per 1M tokens. Output prices run from $0.10 to $3.60.

Context
1,000,000
Max output
128,000
Cache read
$0.15
Cache write
—
Reasoning
Yes
Tool calls
Yes
Structured output
Unknown
Vision
No
Open weights
Yes
License
—
Released
Jun 4, 2026
Updated in source
Jun 4, 2026
Knowledge
—
Providers
22

What this model costs per month

Prices are USD per 1M tokens. A month is 30 days. Cache hits use the cache-read rate when the model has one; otherwise the cache field is ignored.

monthly = requests/day × 30 × (input tokens × (1 − cache hit) × input price + input tokens × cache hit × cache-read price + output tokens × output price) ÷ 1,000,000

Providers

Base rates from Models.dev. Where a provider lists the model more than once, the row is the listing whose id is closest to the canonical model id.
ProviderInputOutputCache readCache writeStatus
Ollama Cloud$0.10$3.00$0.10——
routing.run$0.10$0.10———
CoreWeave$0.50$2.15$0.10——
DevPass$0.50$2.20$0.10——
Eden AI$0.50$2.20$0.10——
Kilo Gateway$0.50$2.20$0.10——
NanoGPT$0.50$2.50$0.25——
Nvidiareference$0.50$2.50$0.15——
Pioneer$0.50$2.50$0.15$0.50—
Baseten$0.60$2.40$0.12——
Fireworks AI$0.60$2.40$0.12——
OpenRouter$0.60$2.40$0.12——
Together AI$0.60$3.60$0.20——
Vercel AI Gateway$0.60$2.40$0.12——
Venice AI$0.63$3.13$0.19——
Nebius Token Factory$1.00$3.00$1.00——
AIHubMix$0$0———
Bothub$0$0———
Kenari$0$0———
OpenCode$0$0$0——
Requesty$0$0———
UnoRouter$0$0———

Jobs this model qualifies for