GitHub
Start free
Dashboard
Start freeSign inDashboard

Nemotron 3 Ultra

by NVIDIA · nvidia/nemotron-3-ultra-550b-a55b

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...

Use in KryptonUse in Krypton
Input / 1M tokens
$0.60
Output / 1M tokens
$2.40
Context
262K

Specs

Input price$0.60 / 1M tokens
Output price$2.40 / 1M tokens
Cached input$0.12 / 1M tokens
Context window262K tokens
Max output183K tokens
Readstext
Writestext
Tool callingYes
ReasoningYes
ListedJun 4, 2026

List prices in USD, updated Oct 30, 2026. Providers may charge differently.

Providers

5 providers serve this model, as of Oct 30, 2026, 18:12 UTC.

Providers serving this model, sorted by price
ProviderQuantizationContextMax outputInput / 1MOutput / 1MCache read / 1MUptime (30 min)Latency (median)Throughput (median)SupportsStatus
DeepInfraCheapestfp4262K16.4K$0.40$1.76$0.08—2.86 s30 tok/sNormal
DeepInfrafp4262K16.4K$0.50$2.20$0.10100%887 ms62 tok/sNormal
BaseTenFastestfp4203K183K$0.60$2.40$0.12100%285 ms102 tok/sNormal
BaseTenfp4203K183K$0.60$2.40$0.12100%269 ms95 tok/sNormal
Venicefp8256K32.8K$0.625$3.13$0.188100%1.04 s61 tok/sNormal

"Cheapest" is the lowest input plus output price; "Fastest" the highest median throughput. Columns empty for every provider are not shown.

All models