Noometry

Model comparison

GPT-4 vs Llama 3.1 Nemotron Ultra 253b v1

Llama 3.1 Nemotron Ultra 253b v1 is the stronger model overall, scoring 36.7 to 29.1 on the Noometry Index.

Last verified . 10 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Llama 3.1 Nemotron Ultra 253b v1 NVIDIA

36.7

Rank #213 Confirmed

Summary

  • They share 10 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Llama 3.1 Nemotron Ultra 253b v1 in 7 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where Llama 3.1 Nemotron Ultra 253b v1 leads 37.5 to 10.8.
  • Llama 3.1 Nemotron Ultra 253b v1 has downloadable open weights; the other is API-only.

Side by side

GPT-4Llama 3.1 Nemotron Ultra 253b v1
ProviderOpenAINVIDIA
Noometry Index29.136.7
Released2023-03-14—
WeightsProprietaryOpen
Context window8K—
Max output8K—
Input $ / M tokens$30—
Output $ / M tokens$60—
Results tracked3811

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

Category by category

Coding Llama 3.1 Nemotron Ultra 253b v1 leads

GPT-4: 31.6 (#283), Llama 3.1 Nemotron Ultra 253b v1: 38.4 (#177)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
LMArena Coding12541312
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Llama 3.1 Nemotron Ultra 253b v1: 15.7 (#149)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
Berkeley Function Calling Leaderboard—10%
METR Time Horizons36.1%—

Reasoning Llama 3.1 Nemotron Ultra 253b v1 leads

GPT-4: 17.8 (#289), Llama 3.1 Nemotron Ultra 253b v1: 26.3 (#134)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
LMArena Hard Prompts12411316
Chess Puzzles4%—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
Epoch Capabilities Index125.89—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Llama 3.1 Nemotron Ultra 253b v1 leads

GPT-4: 10.8 (#309), Llama 3.1 Nemotron Ultra 253b v1: 37.5 (#152)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
LMArena Math12691360
OTIS Mock AIME 2024-20251.1%—
MATH Level 523%—
GSM8K92%—

Knowledge Not comparable

GPT-4: 18.4 (#282), Llama 3.1 Nemotron Ultra 253b v1: —

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
GPQA Diamond35.7%—
LMArena Expert1211—
MMLU86.4%—
TriviaQA84.8%—

Multilingual Llama 3.1 Nemotron Ultra 253b v1 leads

GPT-4: 40.6 (#215), Llama 3.1 Nemotron Ultra 253b v1: 43.1 (#187)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
LMArena Non-English12461282
LMArena Russian12511284
LMArena Chinese1242—
LMArena French1283—
LMArena German1251—
LMArena Japanese1209—
LMArena Korean1184—
LMArena Spanish1261—

Instruction Following Llama 3.1 Nemotron Ultra 253b v1 leads

GPT-4: 65.3 (#222), Llama 3.1 Nemotron Ultra 253b v1: 69.0 (#178)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
LMArena Instruction Following12411308

Long Context Llama 3.1 Nemotron Ultra 253b v1 leads

GPT-4: 37.7 (#212), Llama 3.1 Nemotron Ultra 253b v1: 39.5 (#177)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
LMArena Longer Query12441299

Writing & Preference Llama 3.1 Nemotron Ultra 253b v1 leads

GPT-4: 34.9 (#268), Llama 3.1 Nemotron Ultra 253b v1: 52.2 (#175)

BenchmarkGPT-4Llama 3.1 Nemotron Ultra 253b v1
LMArena Text12631320
LMArena Creative Writing12441314
LMArena Multi-Turn12571317
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than Llama 3.1 Nemotron Ultra 253b v1?

Llama 3.1 Nemotron Ultra 253b v1 is the stronger model overall, scoring 36.7 to 29.1 on the Noometry Index.

Is GPT-4 or Llama 3.1 Nemotron Ultra 253b v1 better for coding?

Llama 3.1 Nemotron Ultra 253b v1 scores higher on coding benchmarks: 38.4 versus 31.6 in the Noometry coding category.

How many benchmarks do GPT-4 and Llama 3.1 Nemotron Ultra 253b v1 share?

10 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Llama 3.1 Nemotron Ultra 253b v1 has 11.

Related comparisons

Go deeper