llm.ing

nemotron-3-nano-30b-a3b

Nvidiaopen weightsOpenRouter ↗Hugging Face ↗Artificial Analysis ↗

Small Nemotron 3 MoE for efficient coding, math, and long-context agents

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Agentic Indexagentic1.0mirroropenrouterSep 8, 2026
AA Coding Indexcoding14.4mirroropenrouterJul 23, 2026
AA Intelligence Indexintelligence8.9mirroropenrouterSep 8, 2026
6.8independentaaSep 9, 2026
LMArena Textpreference1348crowdlmarenaSep 13, 2026
Median output speedspeed160 tok/sindependentaaSep 21, 2026
Median time to first tokenspeed0.94sindependentaaSep 21, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
Crusoefp8262K$0.05 · $0.2100.00%
DeepInfrafp4262K$0.05 · $0.299.53%
Nebiusfp8262K$0.06 · $0.2499.07%
Novitafp4262K$0.05 · $0.299.98%