LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Nemotron 3 Ultra 550B A55B

nvidia·nvidia/nemotron-3-ultra-550b-a55b·GA·Open weights·nemotron series
Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy

Specs & pricing

Input / output per 1M tokens
Official price·Nvidia
$0 / $0
Blended $0 · Cache read —
Lowest paid·routing.runGateway
$0.10 / $0.10
Blended $0.10 · 18× spread
10 more $0 channels
Context
1,000,000
Output limit
65,536
Knowledge cutoff
2025-09
Released / updated
2026-06-04 / 2026-06-04
Capabilities
✓ Reasoning✓ Tool useStructured output✓ TemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
Text

Nemotron 3 Ultra 550B A55B by nvidia is offered by 34 providers on this page. Public prices are shown for 24 of them. Its official list price is $0 per 1M tokens. No paid channel is priced below the official list price. Prices vary about 18× across the providers that host it. It also has 10 free ($0) channels; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.

Artificial Analysis rates it 11.8 on the Intelligence Index, with 24.2 for Finance & Accounting, its strongest domain. Against its lowest blended price of $0.10 per 1M, that is roughly 118 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 255 tokens per second, with 1.03s to the first token. Running one task of the Artificial Analysis suite costs about $2.1897, which reflects how many tokens its reasoning consumes rather than the unit price alone.

The context window is 1,000,000 tokens at the reference host, but hosts report different limits, from 65,536 to 1,048,576, so the usable window depends on the provider you pick. It supports reasoning and tool use. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2025-09.

Quality & performanceArtificial Analysis · Intelligence Index v5 · rep. tier Reasoning

Intelligence11.8
Value118
Engineering22.8
Finance & Accounting24.2
Legal22.6
Healthcare & Medical18
Strategy & Ops15.3
Economics24.1
Output speed255 tok/s
TTFT1.03 s
Cost per task$2.19
Value formulaIQ 11.8 ÷ min blended $0.100 = 118

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 34 providers24 with public prices · 10 free

  • ProviderInput / Output
  • NvidiaOfficialCloud
    Free
    Context 1,000,000Output limit 65,536
  • RequestyGateway
    Free
    Context 1,048,576 ⚠Output limit 65,536
  • OpenRouterGateway
    nvidia/nemotron-3-ultra-550b-a55b:free
    Free
    Context 1,000,000Output limit 65,536
  • BothubGateway
    nemotron-3-ultra-550b-a55b:free
    Free
    Context 1,000,000Output limit 128,000
  • AIHubMixGateway
    nemotron-3-ultra-550b-a55b-free
    Free
    Context 1,000,000Output limit 128,000
  • Kilo GatewayGateway
    nvidia/nemotron-3-ultra-550b-a55b:free
    Free
    Context 1,000,000Output limit 65,536
  • KenariGateway
    nemotron-3-ultra-550b-a55b:free
    Free
    Context 1,000,000Output limit 128,000
  • InferAllGateway
    Free
    Context 1,000,000Output limit 128,000
  • UnoRouterGateway
    nemotron-3-ultra-550b-a55b:free
    Free
    Context 1,000,000Output limit 128,000
  • OpenCodeGateway
    nemotron-3-ultra-free
    Free
    Cache read $0Context 1,000,000Output limit 128,000deprecated
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
NvidiaOfficialCloudFree——1,000,00065,536
RequestyGatewayFree——1,048,576 ⚠65,536
OpenRouter
nvidia/nemotron-3-ultra-550b-a55b:free
GatewayFree——1,000,00065,536
Bothub
nemotron-3-ultra-550b-a55b:free
GatewayFree——1,000,000128,000
AIHubMix
nemotron-3-ultra-550b-a55b-free
GatewayFree——1,000,000128,000
Kilo Gateway
nvidia/nemotron-3-ultra-550b-a55b:free
GatewayFree——1,000,00065,536
Kenari
nemotron-3-ultra-550b-a55b:free
GatewayFree——1,000,000128,000
InferAllGatewayFree——1,000,000128,000
UnoRouter
nemotron-3-ultra-550b-a55b:free
GatewayFree——1,000,000128,000
OpenCode
nemotron-3-ultra-free
GatewayFree$0—1,000,000128,000deprecated

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

effort = noneeffort = mediumeffort = highbudget_tokenseffort = loweffort = maxToggle (on / off)

Interleaved thinking (reasoning between tool calls) is declared by 6 of 34 providers.

7 / 34 providers expose no reasoning control (reasoning_options: []).

Your usage cost

Common scenarios
1routing.run$25.00
2CoreWeave$159.50
3DevPass (LLM Gateway)$162.00
4OpenRouter$162.00
5Eden AI$162.00
6Kilo Gateway$162.00

10 more channels offer $0 (Nvidia, Requesty, OpenRouter etc.); free tiers usually have rate limits and no SLA, excluded from ranking.

The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$2.50$02026-08-052026-10-22
2026-10-22Input list $0Output list $0Min blended $0.10

Benchmark11 items

NameConditionsScoreMetricSource
SWE-Bench Verified—70.7resolvedSource ↗
SWE-Bench Multilingual—67.7resolve rateSource ↗
Terminal-Benchv2.156.4success rateSource ↗
GPQAvariant: no tools87accuracySource ↗
Humanity's Last Examvariant: no tools26.7accuracySource ↗
Humanity's Last Examvariant: with tools37.4accuracySource ↗
LiveCodeBenchvv689pass@1Source ↗
MMLU-Pro—86.8accuracySource ↗
BrowseComp—44.4accuracySource ↗
IFBenchvariant: prompt loose81.7accuracySource ↗
GDPval—46.7wins or tiesSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Artificial Analysis evaluations28 items

GPQA Diamond86.7%
Humanity's Last Exam28.4%
HLE-Diamond8.5%
CritPt3.1%
CritPt v2 Hard15.0%
SciCode40.3%
FrontierCode10.5%
FrontierCode 2.01.1%
Terminal-Bench Hard36.4%
Terminal-Bench 2.153.9%
Terminal-Bench 4.00.5%
Terminal-Bench-Science0.0%
τ²-Bench Telecom83.3%
τ³-Bench Banking14.2%
AutomationBench-AA3.0%
AA-AnalystAgent6.3%
EnterpriseOps-Gym-AA28.9%
AA-Briefcase877
GDPval-AA1021
GDP.pdf5.0%
AA-LCR79.3%
MLCR-AA5.6%
IFBench81.4%
AA-Omniscience Index3.4
AA-Omniscience Accuracy21.2%
AA-Omniscience Non-Hallucination77.4%
AA-3D583
Code2Video-AA304

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and scale (percent correct, points or an index), so rows are not comparable with one another.

Related models

Nemotron 3.5 Super VL 120B A12Bsame series$0 / $0Nemotron 3.5 Super 120B A12Bsame series—Nemotron 3.5 Lightning 30B A3Bsame series$0 / $0Nemotron 3 Super 120B A12Bsame lab—Nemotron 3.5 Content Safetysame lab$0 / $0
Data partly from models.dev (MIT) · Nvidia official docs ↗