llm.ing

Nemotron 3 Ultra 550B A55B

Nvidiaopen weightsOpenRouter ↗Hugging Face ↗

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

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Agentic Indexagentic20.1—mirroropenrouterSep 20, 2026
AA Coding Indexcoding49.3—mirroropenrouterJul 23, 2026
AA Intelligence Indexintelligence11.8—mirroropenrouterOct 28, 2026
Design Arena — 3dpreference1133—mirroropenrouterOct 16, 2026
1102—mirroropenrouterNov 19, 2026
Design Arena — asciiartpreference1096—mirroropenrouterOct 16, 2026
1099—mirroropenrouterOct 17, 2026
1080—mirroropenrouterNov 19, 2026
Design Arena — codecategoriespreference1149—mirroropenrouterOct 16, 2026
1148—mirroropenrouterOct 16, 2026
1127—mirroropenrouterNov 18, 2026
Design Arena — datavizpreference1143—mirroropenrouterOct 16, 2026
1142—mirroropenrouterOct 17, 2026
1136—mirroropenrouterNov 19, 2026
Design Arena — gamedevpreference1140—mirroropenrouterOct 17, 2026
1119—mirroropenrouterNov 19, 2026
Design Arena — pdfpreference1149—mirroropenrouterNov 19, 2026
Design Arena — svgpreference1068—mirroropenrouterOct 17, 2026
1056—mirroropenrouterNov 19, 2026
Design Arena — uicomponentpreference1140—mirroropenrouterOct 17, 2026
1123—mirroropenrouterNov 19, 2026
Design Arena — websitepreference1145—mirroropenrouterOct 16, 2026
1126—mirroropenrouterNov 18, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
BaseTenfp4203K$0.6 · $2.499.91%——
DeepInfrafp4262K$0.5 · $2.299.55%——
DeepInfrafp4262K$0.4 · $1.7682.54%——
DeepInfrafp4262K$0.75 · $3.395.24%——
KT—262K$0.5 · $2.2100.00%——
Venicefp8256K$0.625 · $3.1397.77%——