N
Nemotron 3 Ultra
NVIDIA's 550B/55B open flagship with 1M context, tied into the NIM ecosystem
Intelligence v5
— v4.3.2 —
Leaderboard rank
—
Input / output
$0.40/$1.60
Blended 3:1
$0.70
Output speed
70 tok/s
Value
—
Capability profile Normalised to the field maximum
Specifications
Vendor
NVIDIA
Released
2026-06
Weights
Open (OpenMDW-1.1)
Context
1,000,000 tokens
Max output
Undisclosed
Parameters
550B total / 55B active
Input price
$0.40 / 1M
Output price
$1.60 / 1M
Cache hit
n/a
Output speed
70 tok/s
Intelligence
not indexed
Benchmark scores
How it is measured →Bars are scaled to the field maximum for that benchmark.
| Benchmarks | Score | Vs. best in field | Gap |
|---|---|---|---|
| AA Intelligence Index v5 | — | — | not indexed |
| AA Intelligence Index v4.3.2 | — | — | not indexed |
| GPQA Diamond | — | — | not indexed |
| Terminal-Bench 4.0 | — | — | not indexed |
| LiveBench Design Document generation (DOCX/PPT/PDF) | 62.8 | 72% | −24.1 pts |
| Vals Index | — | — | not indexed |
Assessment
OpenMDW-1.1 licence (not OSI-approved, so read the terms before commercial use). LiveBench design 62.8 trails same-size models because the training data leans technical. Its strength is deep integration with NVIDIA NIM microservices.
At the same price
The four models closest on blended price, side by side.
| Model | v5 | Blended | tok/s | Score gap |
|---|---|---|---|---|
| MiMo-V2.6 Pro Open weightsXiaomi | 30.0 | $0.61 | 81 | — |
| DeepSeek V4.1 Flash Open weightsDeepSeek | 32.5 | $0.53 | 148 | — |
| GLM-5.3 Open weightsZ.ai | 32.0 | $1.00 | 77 | — |
| MiniMax M3.1 FlashMiniMax | 29.0 | $0.26 | 118 | — |
Compare side by side
Line it up against other models on price, speed and every benchmark.
The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.
Open the comparison tool → Back to leaderboard