nemotron-3-ultra-550b-a55b

1427 Score third-party Elo · arena.ai / LMArena ?

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...

LLM & Generative AI 1 model 3 listings snapshot Oct 14, 2026, 12:53 AM

Platforms

click a card to open the detail table

Cheapest OpenRouter no official benchmark Input$0 Output$0 per 1M tokens Fastest OpenRouter lowest p50 latency on this board 128.1 ms p50 Most free quota OpenRouter · nvidia/nemotron-3-ultra-550b-a55b:free free listing ($0 per 1M) $0 free tier

Model variants

W3r9 dedup — same model + identical tags merge; prices are listed per billing form

Model	Provider	Tags	Input /1M	Output /1M
nvidia/nemotron-3-ultra-550b-a55b	OpenRouter	550b256k contextreasoning	$0.600	$2.40
nvidia/nemotron-3-ultra-550b-a55b:free :free	OpenRouter	550b1M contextreasoning	$0	$0
nvidia/nemotron-3-ultra-550b-a55b	Together	550b500k context	$0.600	$3.60

Capabilities

official design_arena & architecture metadata

Design Arena Elo official design_arena benchmark (OpenRouter)

codecategories 1148 · 36.0% wr
website 1145 · 35.0% wr
gamedev 1142 · 36.3% wr
uicomponent 1142 · 36.2% wr
dataviz 1141 · 37.3% wr
3d 1132 · 39.0% wr
asciiart 1097 · 37.5% wr
svg 1068 · 32.4% wr

Input / Output modalities modality: text->text

text in+out

Platform details

per provider

provider	model	Input /1M	Output /1M	Latency ?	Official $/1M	Availability	Data time
DeepInfra	nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B	$0.500	$2.20	—	—	—	2026-10-14 00:43:44
OpenRouter	nvidia/nemotron-3-ultra-550b-a55b	$0.600	$2.40	128.1 ms	—	100% available	2026-10-14 00:43:21
Together	nvidia/nemotron-3-ultra-550b-a55b	$0.600	$3.60	—	—	—	2026-10-14 00:53:13

3 rows · snapshot 2026-10-14 00:53:13 · from data/public/prices.json + measurements.json