AI Battle > Rankings > Nemotron 3 Ultra 550B A55B (Reasoning)

Nemotron 3 Ultra 550B A55B (Reasoning) — Placements in 46 Rankings (2026)

Nemotron 3 Ultra 550B A55B (Reasoning), published by NVIDIA, appears in 46 of the 129 datasets AI Battle tracks. Its strongest published placement is #1 of 12 in Artificial Analysis Openness Index: Score.

Last updated: . Every placement below comes from the named public dataset; AI Battle joins them by slug and changes nothing else.

All placements for Nemotron 3 Ultra 550B A55B (Reasoning)

RankingRankValue
Artificial Analysis Openness Index: Score#1 of 1283.3
Artificial Analysis Openness Index: Components#1 of 126.00
𝜏³-Banking: Time per Task#2 of 200.72
Time per Intelligence Index Task#3 of 202.16
Time per Intelligence Index Task#3 of 202.16
Terminal-Bench v4.0: Time per Task#3 of 203.93
AA-Omniscience Index: Cost Breakdown#3 of 200.47
End-to-End Response Time#4 of 201.83
Latency: Time To First Answer Token#4 of 201.83
AA-Briefcase Output Tokens per Task#4 of 2043461
Time per Task#4 of 204.79
GDPval-AA v2: Output Tokens per Task#4 of 2023545
SciCode: Time per Task#4 of 200.30
AA-Omniscience Hallucination Rate#5 of 200.30
SciCode: Cost per Task#5 of 200.00
Output Tokens per Intelligence Index Task#6 of 2013316
Output Speed#6 of 20204
Output Speed#6 of 20204
Output Tokens per Intelligence Index Task#6 of 2013316
Terminal-Bench v4.0: Output Tokens per Task#6 of 2034172
AA-Omniscience Index: Token Usage#7 of 20788427
SciCode: Output Tokens per Task#7 of 20743
Cost to Run Artificial Analysis Intelligence Index#8 of 2044.8
Pricing: Cache Hit, Input, and Output#8 of 200.60
Cost to Run Artificial Analysis Intelligence Index#8 of 2044.8
Pricing: Cache Hit, Input, and Output#8 of 200.60
Humanity's Last Exam: Cost per Task#8 of 200.00
Cost per Intelligence Index Task#9 of 200.01
Cost per Intelligence Index Task#9 of 200.01
Terminal-Bench v4.0: Cost per Task#9 of 200.00
AA-LCR v1.1: Cost per Task#9 of 200.06
GDPval-AA v2: Cost per Task#9 of 200.00
𝜏³-Banking: Output Tokens per Task#9 of 203294
Model Size: Total and Active Parameters#7 of 1455.0
AA-Briefcase Cost per Task#10 of 200.10
Humanity's Last Exam: Time per Task#11 of 202.83
Artificial Analysis Openness Index: Components#12 of 206.00
Artificial Analysis Openness Index: Score#12 of 2083.3
𝜏³-Banking: Cost per Task#12 of 200.00
AA-LCR v1.1: Time per Task#15 of 200.33
AA-Omniscience Index#18 of 20-0.40
AA-Omniscience Index#18 of 20-0.40
AA-Briefcase Elo#20 of 20878
Artificial Analysis Finance & Accounting Index#20 of 2024.7
AA-Briefcase Elo#20 of 20878
AA-AnalystAgent pass^5#11 of 110.06

Price vs intelligence

intelligence index →input price / 1M → $4.00
Each dot is a model, joined by AI Battle from two public datasets — the intelligence index ranking and the published input pricing ranking. Every published value is shown as-is, no averaging. The dashed line is the value frontier: no model above or to the right of it offers a lower input price at its intelligence level.

How to read this profile

Each row is a separate measurement in a separate dataset — the position reflects the metric of that ranking, and a higher number is not "better" for every dataset (times and prices are costs). AI Battle deliberately does not average placements into a single score: datasets differ in coverage, so a total would hide more than it shows.

Raw data

Machine-readable profile: /data/models/nvidia-nemotron-3-ultra-550b-a55b.json.

Method

Placements, joins and this page are computed by AI Battle from public datasets — see the methodology for sources, limitations and corrections.