Noometry

Model comparison

Claude 3 Haiku vs Nemotron 3 Ultra

Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 25.9 on the Noometry Index.

Last verified . 23 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Nemotron 3 Ultra NVIDIA

42.5

Rank #113 Confirmed

Summary

  • They share 23 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Nemotron 3 Ultra in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Nemotron 3 Ultra leads 66.3 to 29.7.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 86.7% for Nemotron 3 Ultra.
  • Nemotron 3 Ultra has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuNemotron 3 Ultra
ProviderAnthropicNVIDIA
Noometry Index25.942.5
Released2024-03-072026-06-04
WeightsProprietaryOpen
Context window—262K
Max output—128K
Input $ / M tokens—$0.50
Output $ / M tokens—$2.20
Results tracked3730

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Nemotron 3 Ultra leads

Claude 3 Haiku: 26.4 (#325), Nemotron 3 Ultra: 38.1 (#182)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
WeirdML9.8%43.5%
LMArena Coding11991468
FrontierCode—13.6%
SciCode—40.3%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, Nemotron 3 Ultra: 23.4 (#126)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
APEX-Agents—22.7%

Reasoning Nemotron 3 Ultra leads

Claude 3 Haiku: 16.3 (#307), Nemotron 3 Ultra: 31.0 (#82)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
LMArena Hard Prompts11741452
DTBench50.1%90.1%
LMCA8.8%36.9%
Epoch Capabilities Index118.35146.17
Kagi LLM Benchmark34.2%—
CritPt—3.1%
Chess Puzzles—12%
Mystery Game Puzzles—20%
ForecastBench53.2—
WinoGrande74.2%—

Math Nemotron 3 Ultra leads

Claude 3 Haiku: 9.8 (#319), Nemotron 3 Ultra: 35.3 (#187)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
OTIS Mock AIME 2024-20251.8%86.7%
LMArena Math11881457
ProofBench—2%
MATH Level 514.9%—

Knowledge Nemotron 3 Ultra leads

Claude 3 Haiku: 17.3 (#285), Nemotron 3 Ultra: 52.5 (#61)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
GPQA Diamond36.3%85.4%
LMArena Expert11481472
Confabulations34.2%—
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), Nemotron 3 Ultra: —

BenchmarkClaude 3 HaikuNemotron 3 Ultra
LMArena Vision950—
ScienceQA72%—

Multilingual Nemotron 3 Ultra leads

Claude 3 Haiku: 36.0 (#243), Nemotron 3 Ultra: 53.5 (#64)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
LMArena Non-English11781427
LMArena Chinese11551497
LMArena French11951472
LMArena German11741471
LMArena Korean11091386
LMArena Russian12041417
LMArena Spanish11661454
LMArena Japanese1102—

Instruction Following Nemotron 3 Ultra leads

Claude 3 Haiku: 61.3 (#247), Nemotron 3 Ultra: 74.7 (#88)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
LMArena Instruction Following11731418

Long Context Nemotron 3 Ultra leads

Claude 3 Haiku: 36.1 (#237), Nemotron 3 Ultra: 43.9 (#81)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
LMArena Longer Query11901437

Writing & Preference Nemotron 3 Ultra leads

Claude 3 Haiku: 29.7 (#291), Nemotron 3 Ultra: 66.3 (#36)

BenchmarkClaude 3 HaikuNemotron 3 Ultra
LMArena Text11951445
LMArena Creative Writing11571398
EQ-Bench Creative Writing7171692
LMArena Multi-Turn11901411

Frequently asked questions

Is Claude 3 Haiku better than Nemotron 3 Ultra?

Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or Nemotron 3 Ultra better for coding?

Nemotron 3 Ultra scores higher on coding benchmarks: 38.1 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Nemotron 3 Ultra share?

23 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Nemotron 3 Ultra has 30.

Related comparisons

Go deeper