Noometry

Model comparison

GPT-5.4 mini vs Llama 2-70B

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 24.4 on the Noometry Index.

Last verified . 22 shared benchmarks.

GPT-5.4 mini OpenAI

45.0

Rank #76 Confirmed

Llama 2-70B Meta

24.4

Rank #349 Confirmed

Summary

  • They share 22 benchmarks with published results for both. GPT-5.4 mini scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where GPT-5.4 mini leads 51.5 to 7.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for GPT-5.4 mini and 0% for Llama 2-70B.
  • Llama 2-70B has downloadable open weights; the other is API-only.

Side by side

GPT-5.4 miniLlama 2-70B
ProviderOpenAIMeta
Noometry Index45.024.4
Released2026-03-172023-07-18
WeightsProprietaryOpen
Context window400K—
Max output128K—
Input $ / M tokens$0.75—
Output $ / M tokens$4.50—
Results tracked4635

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

Category by category

Coding GPT-5.4 mini leads

GPT-5.4 mini: 45.2 (#72), Llama 2-70B: 31.4 (#286)

BenchmarkGPT-5.4 miniLlama 2-70B
LMArena Coding14381079
FrontierCode27%—
LMArena WebDev1397—
SciCode49.9%—
WeirdML60.3%—
ALE-Bench1,189—

Agentic & Tool Use Not comparable

GPT-5.4 mini: 29.9 (#81), Llama 2-70B: —

BenchmarkGPT-5.4 miniLlama 2-70B
DeepResearch Bench36.3%—

Reasoning GPT-5.4 mini leads

GPT-5.4 mini: 30.4 (#85), Llama 2-70B: 14.4 (#325)

BenchmarkGPT-5.4 miniLlama 2-70B
LMArena Hard Prompts14241073
DTBench80%41.6%
Epoch Capabilities Index148.84113.79
ForecastBench5751.4
ARC-AGI-218.9%—
Kagi LLM Benchmark37.9%—
NYT Connections (extended)61.8%—
ARC-AGI-163.7%—
CritPt10%—
Chess Puzzles24%—
Thematic Generalization61.7%—
Mystery Game Puzzles11%—
LMCA40.8%—
BIG-Bench Hard—64.9%
CommonsenseQA 2.0—50%
HellaSwag—85.3%
LAMBADA—78.9%
PIQA—82.8%
WinoGrande—80.2%

Math GPT-5.4 mini leads

GPT-5.4 mini: 45.5 (#75), Llama 2-70B: 8.1 (#326)

BenchmarkGPT-5.4 miniLlama 2-70B
OTIS Mock AIME 2024-202588.9%0%
LMArena Math14191091
FrontierMath (Tiers 1-3)51.2%—
FrontierMath Tier 49.8%—
ProofBench21%—
MATH Level 5—3.3%
FrontierMath (Feb 2025 set)28.3%—
FrontierMath Tier 4 (v1)2.1%—
GSM8K—69.6%

Knowledge GPT-5.4 mini leads

GPT-5.4 mini: 51.5 (#67), Llama 2-70B: 7.4 (#310)

BenchmarkGPT-5.4 miniLlama 2-70B
GPQA Diamond86.9%26.3%
LMArena Expert14351039
SimpleQA Verified29.4%—
Vectara Hallucination Rate5.5%—
ARC (AI2) Challenge—78.3%
BoolQ—88.6%
MMLU—69.9%
OpenBookQA—60.2%
TriviaQA—87.6%

Multimodal Not comparable

GPT-5.4 mini: 39.7 (#56), Llama 2-70B: —

BenchmarkGPT-5.4 miniLlama 2-70B
LMArena Vision1245—

Multilingual GPT-5.4 mini leads

GPT-5.4 mini: 51.9 (#96), Llama 2-70B: 27.7 (#274)

BenchmarkGPT-5.4 miniLlama 2-70B
LMArena Non-English14051045
LMArena Chinese1446995
LMArena French14401090
LMArena German14091041
LMArena Japanese1374927
LMArena Korean1368964
LMArena Russian14171083
LMArena Spanish14051143

Instruction Following GPT-5.4 mini leads

GPT-5.4 mini: 74.1 (#102), Llama 2-70B: 54.9 (#278)

BenchmarkGPT-5.4 miniLlama 2-70B
LMArena Instruction Following14051071

Long Context GPT-5.4 mini leads

GPT-5.4 mini: 43.0 (#112), Llama 2-70B: 32.3 (#270)

BenchmarkGPT-5.4 miniLlama 2-70B
LMArena Longer Query14071062

Writing & Preference GPT-5.4 mini leads

GPT-5.4 mini: 64.0 (#58), Llama 2-70B: 32.3 (#279)

BenchmarkGPT-5.4 miniLlama 2-70B
LMArena Text14121115
LMArena Creative Writing13701075
LMArena Multi-Turn14291088
EQ-Bench Creative Writing1665—

Frequently asked questions

Is GPT-5.4 mini better than Llama 2-70B?

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 24.4 on the Noometry Index.

Is GPT-5.4 mini or Llama 2-70B better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 31.4 in the Noometry coding category.

How many benchmarks do GPT-5.4 mini and Llama 2-70B share?

22 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Llama 2-70B has 35.

Related comparisons

Go deeper