modelbenchmark.io

Nemotron 3 Ultra 550B A55B (Nebius)

NVIDIA · nemotron-3-ultra-550b-a55b

Compare

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

Specification

most-agreed values

Context
1M
Max output
1M
Released
2026-06-04
Knowledge cutoff
2025-09
Retires
2026-08-27
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$1.00
Output
$3.00
Cache read
$1.00

Available from 19 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
bothub · freefreefree——1M128K—
kilo · freefreefree——1M66K—
nvidiafreefree——1M66K—
openrouter · freefreefree——1M66K—
requestyfreefree——1M66K—
unorouter · freefreefree——1M128K—
perplexity$0.25$2.50$0.25————
kilo$0.40$1.76$0.08—203K183K—
nano-gpt$0.50$2.50$0.15—512K66K—
nano-gpt · thinking$0.50$2.50$0.15—1M66K—
commandcode$0.60$2.40$0.12—1M128K—
edenai · arkanecloud$0.60$3.50$0.20—512K128K—
openrouter$0.60$2.40$0.12—262K183K—
together_ai$0.60$3.60$0.20—512K512K2026-08-27
togetherai$0.60$3.60$0.20—512K512K—
vercel$0.60$2.40$0.12—1M65K—
kenari$0.6857$4.2857$0.0686—1M128K—
edenai
$1.00
$0.50
$3.00
$2.20
$1.00
$0.10
—
1M
262K
128K—
nebius$1.00$3.00$0.12—1M1M—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02edenaiCache read$0.10
2026-09-04edenaiCache read$1.00
2026-09-24edenaiCache read$0.10
2026-10-19edenaiCache read$1.00
2026-10-20edenaiCache read$0.10
2026-10-22edenaiCache read$1.00
2026-10-23edenaiCache read$0.10
2026-10-25edenaiCache read$1.00
2026-10-26edenaiCache read$0.10
2026-10-28edenaiCache read$1.00
2026-10-31edenaiCache read$0.10
2026-11-01edenaiCache read$1.00
2026-09-02kiloCache read$0.10
2026-10-16kiloCache read$0.08
2026-09-02nano-gptCache read$0.25
2026-10-30nano-gptCache read$0.15
2026-09-02nano-gpt · thinkingCache read$0.25
2026-10-30nano-gpt · thinkingCache read$0.15
2026-09-02openrouterCache read$0.10
2026-09-02openrouterCache read$0.1875
2026-09-03openrouterCache read$0.12
2026-09-04openrouterCache read$0.1875
2026-09-06openrouterCache read$0.1875
2026-09-14openrouterCache read$0.12
2026-09-16openrouterCache read$0.1875
2026-09-19openrouterCache read$0.12
2026-09-20openrouterCache read$0.12
2026-10-02openrouterCache read$0.10
2026-10-03openrouterCache read$0.10
2026-10-10openrouterCache read$0.10
2026-10-12openrouterCache read$0.12
2026-09-02together_aiCache read$0.20
2026-09-24together_aiCache read$0.20
2026-09-02edenaiInput$0.50
2026-09-04edenaiInput$1.00
2026-09-24edenaiInput$0.50
2026-10-19edenaiInput$1.00
2026-10-20edenaiInput$0.50
2026-10-22edenaiInput$1.00
2026-10-23edenaiInput$0.50
2026-10-25edenaiInput$1.00
2026-10-26edenaiInput$0.50
2026-10-28edenaiInput$1.00
2026-10-31edenaiInput$0.50
2026-11-01edenaiInput$1.00
2026-09-02kenariInputfree
2026-10-22kenariInput$0.6857
2026-09-02kiloInput$0.50
2026-10-16kiloInput$0.40
2026-09-02nvidiaInput$0.50
2026-10-22nvidiaInputfree
2026-09-02openrouterInput$0.50
2026-09-02openrouterInput$0.625
2026-09-03openrouterInput$0.60
2026-09-04openrouterInput$0.625
2026-09-06openrouterInput$0.625
2026-09-14openrouterInput$0.60
2026-09-16openrouterInput$0.625
2026-09-19openrouterInput$0.60
2026-09-20openrouterInput$0.60
2026-10-02openrouterInput$0.50
2026-10-03openrouterInput$0.50
2026-10-10openrouterInput$0.50
2026-10-12openrouterInput$0.60
2026-09-02openrouter · freeInputfree
2026-09-06openrouter · freeInputfree
2026-10-10openrouter · freeInputfree
2026-09-02together_aiInput$0.60
2026-09-24together_aiInput$0.60
2026-09-02edenaiOutput$2.20
2026-09-04edenaiOutput$3.00
2026-09-24edenaiOutput$2.20
2026-10-19edenaiOutput$3.00
2026-10-20edenaiOutput$2.20
2026-10-22edenaiOutput$3.00
2026-10-23edenaiOutput$2.20
2026-10-25edenaiOutput$3.00
2026-10-26edenaiOutput$2.20
2026-10-28edenaiOutput$3.00
2026-10-31edenaiOutput$2.20
2026-11-01edenaiOutput$3.00
2026-09-02kenariOutputfree
2026-10-22kenariOutput$4.2857
2026-09-02kiloOutput$2.20
2026-10-16kiloOutput$1.76
2026-09-02nvidiaOutput$2.50
2026-10-22nvidiaOutputfree
2026-09-02openrouterOutput$2.20
2026-09-02openrouterOutput$3.125
2026-09-03openrouterOutput$2.40
2026-09-04openrouterOutput$3.125
2026-09-06openrouterOutput$3.125
2026-09-14openrouterOutput$2.40
2026-09-16openrouterOutput$3.125
2026-09-19openrouterOutput$2.40
2026-09-20openrouterOutput$2.40
2026-10-02openrouterOutput$2.20
2026-10-03openrouterOutput$2.20
2026-10-10openrouterOutput$2.20
2026-10-12openrouterOutput$2.40
2026-09-02openrouter · freeOutputfree
2026-09-06openrouter · freeOutputfree
2026-10-10openrouter · freeOutputfree
2026-09-02together_aiOutput$3.60
2026-09-24together_aiOutput$3.60
2026-09-03nebiusCache read$1.00
2026-10-24nebiusCache read$0.12
2026-09-03nebiusInput$1.00
2026-09-06nebiusInput$1.00
2026-09-03nebiusOutput$3.00
2026-09-06nebiusOutput$3.00

More from NVIDIA

most-hosted first