Noometry

Model comparison

Llama 3-70B vs o4-mini

o4-mini is the stronger model overall, scoring 41.6 to 28.8 on the Noometry Index.

Last verified . 24 shared benchmarks.

Llama 3-70B Meta

28.8

Rank #323 Confirmed

o4-mini OpenAI

41.6

Rank #132 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Llama 3-70B scores higher in 0 categories and o4-mini in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where o4-mini leads 40.8 to 12.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Llama 3-70B and 81.7% for o4-mini.
  • Llama 3-70B has downloadable open weights; the other is API-only.

Side by side

Llama 3-70Bo4-mini
ProviderMetaOpenAI
Noometry Index28.841.6
Released2024-04-182025-04-16
WeightsOpenProprietary
Context window—200K
Max output—100K
Input $ / M tokens—$1.10
Output $ / M tokens—$4.40
Results tracked3160

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

Category by category

Coding o4-mini leads

Llama 3-70B: 35.8 (#218), o4-mini: 40.9 (#127)

BenchmarkLlama 3-70Bo4-mini
LMArena Coding12061368
SWE-bench Verified (bash only)—45%
Aider Polyglot—72%
GSO—3.6%
WeirdML—52.6%
BigCodeBench Instruct43.6%—
BigCodeBench Complete54.5%—
CadEval—62%
ALE-Bench—826.17
AlgoTune—1.72
HumanEval+72%—
MBPP+69%—

Agentic & Tool Use o4-mini leads

Llama 3-70B: 21.1 (#139), o4-mini: 32.6 (#61)

BenchmarkLlama 3-70Bo4-mini
Berkeley Function Calling Leaderboard—53.2%
GDPval—25.3%
Cybench5%—
METR Time Horizons—63.9%

Reasoning o4-mini leads

Llama 3-70B: 18.0 (#288), o4-mini: 24.6 (#162)

BenchmarkLlama 3-70Bo4-mini
Kagi LLM Benchmark35.1%67.6%
LMArena Hard Prompts11951351
DTBench54.2%77.6%
Epoch Capabilities Index122.93145.64
ForecastBench57.161.8
ARC-AGI-2—6.1%
SimpleBench—38.7%
ARC-AGI-1—58.7%
CritPt—0.6%
Chess Puzzles—26%
EnigmaEval—9.2%
Mystery Game Puzzles—5%
LMCA—26.5%
WinoGrande83.5%—

Math o4-mini leads

Llama 3-70B: 12.8 (#305), o4-mini: 40.8 (#89)

BenchmarkLlama 3-70Bo4-mini
OTIS Mock AIME 2024-20254.3%81.7%
LMArena Math12181389
MATH Level 522.6%97.8%
FrontierMath (Tiers 1-3)—36.1%
FrontierMath Tier 4—4.9%
Omni-MATH—72%
FrontierMath (Feb 2025 set)—24.8%
FrontierMath Tier 4 (v1)—6.3%

Knowledge o4-mini leads

Llama 3-70B: 20.8 (#277), o4-mini: 43.6 (#91)

BenchmarkLlama 3-70Bo4-mini
GPQA Diamond40.6%79.6%
LMArena Expert11491343
Humanity's Last Exam—18.1%
SimpleQA Verified—19.6%
MMLU-Pro—82%
Confabulations—15.8%
Vectara Hallucination Rate—18.6%
GPQA (HELM)—73.5%
MMLU79.3%—

Multimodal Not comparable

Llama 3-70B: —, o4-mini: 40.2 (#49)

BenchmarkLlama 3-70Bo4-mini
LMArena Vision—1194
GeoBench—64%
VPCT—57.5%

Multilingual o4-mini leads

Llama 3-70B: 33.6 (#251), o4-mini: 47.0 (#154)

BenchmarkLlama 3-70Bo4-mini
LMArena Non-English11421337
LMArena Chinese11141354
LMArena French12321364
LMArena German11691336
LMArena Japanese10171308
LMArena Korean10171312
LMArena Russian11591334
LMArena Spanish12411347

Instruction Following o4-mini leads

Llama 3-70B: 62.5 (#238), o4-mini: 75.2 (#68)

BenchmarkLlama 3-70Bo4-mini
LMArena Instruction Following11941321
IFEval—92.8%

Long Context o4-mini leads

Llama 3-70B: 35.6 (#240), o4-mini: 45.5 (#33)

BenchmarkLlama 3-70Bo4-mini
LMArena Longer Query11741315
Fiction.LiveBench—77.8%

Writing & Preference o4-mini leads

Llama 3-70B: 42.8 (#231), o4-mini: 54.0 (#152)

BenchmarkLlama 3-70Bo4-mini
LMArena Text12211353
LMArena Creative Writing12101294
LMArena Multi-Turn12231350
Short-Story Creative Writing—75%
WildBench—85.4%

Frequently asked questions

Is Llama 3-70B better than o4-mini?

o4-mini is the stronger model overall, scoring 41.6 to 28.8 on the Noometry Index.

Is Llama 3-70B or o4-mini better for coding?

o4-mini scores higher on coding benchmarks: 40.9 versus 35.8 in the Noometry coding category.

How many benchmarks do Llama 3-70B and o4-mini share?

24 benchmarks have published results for both models. Llama 3-70B has 31 scored results on Noometry and o4-mini has 60.

Related comparisons

Go deeper