Model record

Llama 3.1 Nemotron Ultra 253B v1

Open weights

NVIDIA · released Apr 7, 2025

Overall standing

Overall
23.8rank 61 of 62
23.8 is at approximately the 2th percentile of the ranked Overall field. The field spans 19.0 to 62.2, with a median of 48.9.
6 of 8 measured · 2 estimated
Category standingsReasoning 31.6Coding —Math 0.0Agentic —
Reasoning
31.6rank 62 of 62
31.6 is at approximately the 1th percentile of the ranked Reasoning field. The field spans 31.6 to 73.3, with a median of 64.8.
4 of 4 measured
Coding
Insufficient data
1 of 2 measured
Math
0.0rank 53 of 62
0.0 is at approximately the 8th percentile of the ranked Math field. The field spans 0.0 to 32.3, with a median of 5.3.
1 of 1 measured
Agentic
Insufficient data
0 of 1 measured
Model factsOpen weights · Price not listed

Copied from the linked provider page; no separate retrieval date is stored.

Provider
NVIDIA
Released
Weights
Open weights
Context window
131,072 tokens
Input price / Mtok
Not listed
Output price / Mtok
Not listed
Reasoning effort
Not listed

Scores and sources

Measured scores retain their source and retrieval date. Missing scores are not treated as zero.

Benchmark scores, evaluation settings, sources, and retrieval dates for Llama 3.1 Nemotron Ultra 253B v1
BenchmarkCategoryScoreEvaluation settingsSource
Terminal-Bench v2.1Coding—Not measured in this datasetNo source
τ³-BankingAgentic—Not measured in this datasetNo source
AA-LCRReasoning7.3Artificial Analysis independent run; 100 roughly 100k-token multi-document questions; 3 repeats; equality-checker pass@1; no toolsOpen source (opens in a new tab)
Humanity's Last ExamReasoning8.1Artificial Analysis independent run; 2,158 text-only questions; pass@1; GPT-4o (August 2024) equality checker; no toolsOpen source (opens in a new tab)
GPQA DiamondReasoning72.8Artificial Analysis independent run; 198 questions; 5 repeats; regex-graded pass@1; no toolsOpen source (opens in a new tab)
SciCodeCoding34.7Artificial Analysis independent run; 288 test subproblems; scientist background included; 3 repeats; subproblem pass@1Open source (opens in a new tab)
IFBenchReasoning38.2Artificial Analysis independent run; 294 single-turn prompts; 5 repeats; official loose evaluator; prompt-level pass@1Open source (opens in a new tab)
CritPtMath0.0Artificial Analysis independent run; 70 test challenges; 5 repeats; two-step answer parsing; official CritPt grader; pass@1; no toolsOpen source (opens in a new tab)