Model comparison

Llama 2-70B vs Phi 3 Mini 4k Instruct June 2024

Phi 3 Mini 4k Instruct June 2024 is the stronger model overall, scoring 31.3 to 24.4 on the Noometry Index.

Last verified . 15 shared benchmarks.

Llama 2-70B Meta

24.4

Rank #349 Confirmed

Summary

  • They share 15 benchmarks with published results for both. Llama 2-70B scores higher in 4 categories and Phi 3 Mini 4k Instruct June 2024 in 4 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in math, where Phi 3 Mini 4k Instruct June 2024 leads 33.0 to 8.1.

Side by side

Llama 2-70B and Phi 3 Mini 4k Instruct June 2024 specifications
Llama 2-70BPhi 3 Mini 4k Instruct June 2024
ProviderMetaMicrosoft
Noometry Index24.431.3
Released2023-07-18—
WeightsOpenOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked3515

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

Category by category

Coding Too close to call

Llama 2-70B: 31.4 (#286), Phi 3 Mini 4k Instruct June 2024: 31.8 (#279)

Coding benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Coding10791093

Reasoning Phi 3 Mini 4k Instruct June 2024 leads

Llama 2-70B: 14.4 (#325), Phi 3 Mini 4k Instruct June 2024: 20.8 (#231)

Reasoning benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Hard Prompts10731087
DTBench41.6%—
BIG-Bench Hard64.9%—
CommonsenseQA 2.050%—
Epoch Capabilities Index113.79—
ForecastBench51.4—
HellaSwag85.3%—
LAMBADA78.9%—
PIQA82.8%—
WinoGrande80.2%—

Math Phi 3 Mini 4k Instruct June 2024 leads

Llama 2-70B: 8.1 (#326), Phi 3 Mini 4k Instruct June 2024: 33.0 (#208)

Math benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Math10911152
OTIS Mock AIME 2024-20250%—
MATH Level 53.3%—
GSM8K69.6%—

Knowledge Phi 3 Mini 4k Instruct June 2024 leads

Llama 2-70B: 7.4 (#310), Phi 3 Mini 4k Instruct June 2024: 28.6 (#244)

Knowledge benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Expert10391051
GPQA Diamond26.3%—
ARC (AI2) Challenge78.3%—
BoolQ88.6%—
MMLU69.9%—
OpenBookQA60.2%—
TriviaQA87.6%—

Multilingual Llama 2-70B leads

Llama 2-70B: 27.7 (#274), Phi 3 Mini 4k Instruct June 2024: 25.9 (#282)

Multilingual benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Non-English10451013
LMArena Chinese9951033
LMArena German10411031
LMArena Japanese927954
LMArena Korean964880
LMArena Russian10831019
LMArena French1090—
LMArena Spanish1143—

Instruction Following Too close to call

Llama 2-70B: 54.9 (#278), Phi 3 Mini 4k Instruct June 2024: 54.1 (#282)

Instruction Following benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Instruction Following10711058

Long Context Too close to call

Llama 2-70B: 32.3 (#270), Phi 3 Mini 4k Instruct June 2024: 31.7 (#277)

Long Context benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Longer Query10621042

Writing & Preference Llama 2-70B leads

Llama 2-70B: 32.3 (#279), Phi 3 Mini 4k Instruct June 2024: 29.6 (#294)

Writing & Preference benchmarks
BenchmarkLlama 2-70BPhi 3 Mini 4k Instruct June 2024
LMArena Text11151080
LMArena Creative Writing10751045
LMArena Multi-Turn10881049

Frequently asked questions

Is Llama 2-70B better than Phi 3 Mini 4k Instruct June 2024?

Phi 3 Mini 4k Instruct June 2024 is the stronger model overall, scoring 31.3 to 24.4 on the Noometry Index.

Is Llama 2-70B or Phi 3 Mini 4k Instruct June 2024 better for coding?

They score almost the same on coding (31.4 vs 31.8); test both on your own repository before choosing.

How many benchmarks do Llama 2-70B and Phi 3 Mini 4k Instruct June 2024 share?

15 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Phi 3 Mini 4k Instruct June 2024 has 15.

Related comparisons

Go deeper