Noometry

Model comparison

GPT-4 vs Llama 3.1-70B

GPT-4 and Llama 3.1-70B score almost the same on the Noometry Index (29.1 vs 29.6), so choose on price, context window or the category you care about most.

Last verified . 28 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Llama 3.1-70B Meta

29.6

Rank #308 Confirmed

Summary

  • They share 28 benchmarks with published results for both. GPT-4 scores higher in 4 categories and Llama 3.1-70B in 4 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Llama 3.1-70B leads 24.2 to 18.4.
  • The biggest single-benchmark swing is MATH Level 5: 23% for GPT-4 and 36.7% for Llama 3.1-70B.
  • Llama 3.1-70B is cheaper at $0.40 / $0.40 per million input/output tokens, against $30 / $60 for GPT-4.
  • Llama 3.1-70B accepts more context: 128K tokens versus 8K.
  • Llama 3.1-70B has downloadable open weights; the other is API-only.

Side by side

GPT-4Llama 3.1-70B
ProviderOpenAIMeta
Noometry Index29.129.6
Released2023-03-142024-07-23
WeightsProprietaryOpen
Context window8K128K
Max output8K4K
Input $ / M tokens$30$0.40
Output $ / M tokens$60$0.40
Results tracked3835

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

Category by category

Coding GPT-4 leads

GPT-4: 31.6 (#283), Llama 3.1-70B: 30.3 (#296)

BenchmarkGPT-4Llama 3.1-70B
WeirdML12.4%9%
BigCodeBench Instruct46%46.1%
LMArena Coding12541260
BigCodeBench Complete57.2%54.8%
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Llama 3.1-70B: 25.1 (#112)

BenchmarkGPT-4Llama 3.1-70B
TheAgentCompany—6.9%
BALROG—27.9%
METR Time Horizons36.1%—

Reasoning Llama 3.1-70B leads

GPT-4: 17.8 (#289), Llama 3.1-70B: 21.6 (#220)

BenchmarkGPT-4Llama 3.1-70B
LMArena Hard Prompts12411241
DTBench62.7%60%
LMCA17.1%14.8%
Epoch Capabilities Index125.89125.92
Chess Puzzles4%—
Mystery Game Puzzles12%—
BIG-Bench Hard75.1%—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Llama 3.1-70B leads

GPT-4: 10.8 (#309), Llama 3.1-70B: 13.5 (#304)

BenchmarkGPT-4Llama 3.1-70B
OTIS Mock AIME 2024-20251.1%3.6%
LMArena Math12691252
MATH Level 523%36.7%
Omni-MATH—21%
GSM8K92%—

Knowledge Llama 3.1-70B leads

GPT-4: 18.4 (#282), Llama 3.1-70B: 24.2 (#269)

BenchmarkGPT-4Llama 3.1-70B
GPQA Diamond35.7%44.2%
LMArena Expert12111209
MMLU86.4%80.1%
MMLU-Pro—65.3%
GPQA (HELM)—42.6%
TriviaQA84.8%—

Multilingual GPT-4 leads

GPT-4: 40.6 (#215), Llama 3.1-70B: 38.8 (#225)

BenchmarkGPT-4Llama 3.1-70B
LMArena Non-English12461219
LMArena Chinese12421215
LMArena French12831261
LMArena German12511222
LMArena Japanese12091132
LMArena Korean11841140
LMArena Russian12511234
LMArena Spanish12611253

Instruction Following Too close to call

GPT-4: 65.3 (#222), Llama 3.1-70B: 65.3 (#223)

BenchmarkGPT-4Llama 3.1-70B
LMArena Instruction Following12411231
IFEval—82.1%

Long Context Too close to call

GPT-4: 37.7 (#212), Llama 3.1-70B: 37.6 (#214)

BenchmarkGPT-4Llama 3.1-70B
LMArena Longer Query12441241

Writing & Preference Too close to call

GPT-4: 34.9 (#268), Llama 3.1-70B: 35.4 (#267)

BenchmarkGPT-4Llama 3.1-70B
LMArena Text12631261
LMArena Creative Writing12441232
EQ-Bench Creative Writing752784
LMArena Multi-Turn12571256
WildBench—75.8%

Frequently asked questions

Is GPT-4 better than Llama 3.1-70B?

GPT-4 and Llama 3.1-70B score almost the same on the Noometry Index (29.1 vs 29.6), so choose on price, context window or the category you care about most.

Which is cheaper, GPT-4 or Llama 3.1-70B?

Llama 3.1-70B is cheaper. It lists at $0.40 per million input tokens and $0.40 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or Llama 3.1-70B better for coding?

GPT-4 scores higher on coding benchmarks: 31.6 versus 30.3 in the Noometry coding category.

Which has the bigger context window?

Llama 3.1-70B does, with 128K tokens against 8K.

How many benchmarks do GPT-4 and Llama 3.1-70B share?

28 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Llama 3.1-70B has 35.

Related comparisons

Go deeper