Noometry

Model comparison

GPT-4 vs Granite 3.1 8b Instruct

Granite 3.1 8b Instruct is the stronger model overall, scoring 31.8 to 28.9 on the Noometry Index.

Last verified . 12 shared benchmarks.

GPT-4 OpenAI

28.9

Rank #308 Confirmed

Granite 3.1 8b Instruct IBM

31.8

Rank #256 Confirmed

Summary

  • They share 12 benchmarks with published results for both. GPT-4 scores higher in 3 categories and Granite 3.1 8b Instruct in 5 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where Granite 3.1 8b Instruct leads 30.4 to 10.1.
  • Granite 3.1 8b Instruct has downloadable open weights; the other is API-only.

Side by side

GPT-4Granite 3.1 8b Instruct
ProviderOpenAIIBM
Noometry Index28.931.8
Released2023-03-14—
WeightsProprietaryOpen
Context window8K—
Max output8K—
Input $ / M tokens$30—
Output $ / M tokens$60—
Results tracked3813

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

Category by category

Coding Granite 3.1 8b Instruct leads

GPT-4: 31.2 (#277), Granite 3.1 8b Instruct: 33.9 (#232)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Coding12541186
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Granite 3.1 8b Instruct: 23.9 (#113)

BenchmarkGPT-4Granite 3.1 8b Instruct
Berkeley Function Calling Leaderboard—27.1%
METR Time Horizons36.1%—

Reasoning Granite 3.1 8b Instruct leads

GPT-4: 17.3 (#285), Granite 3.1 8b Instruct: 21.2 (#204)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Hard Prompts12411146
Chess Puzzles4%—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
Epoch Capabilities Index125.88—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Granite 3.1 8b Instruct leads

GPT-4: 10.1 (#304), Granite 3.1 8b Instruct: 30.4 (#210)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Math12691152
OTIS Mock AIME 2024-20251.1%—
MATH Level 523%—
GSM8K92%—

Knowledge Granite 3.1 8b Instruct leads

GPT-4: 18.4 (#277), Granite 3.1 8b Instruct: 30.6 (#218)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Expert12111142
GPQA Diamond35.7%—
MMLU86.4%—
TriviaQA84.8%—

Multilingual GPT-4 leads

GPT-4: 40.8 (#210), Granite 3.1 8b Instruct: 31.1 (#255)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Non-English12461099
LMArena Chinese12421145
LMArena Russian12511092
LMArena French1282—
LMArena German1251—
LMArena Japanese1209—
LMArena Korean1184—
LMArena Spanish1261—

Instruction Following GPT-4 leads

GPT-4: 65.5 (#217), Granite 3.1 8b Instruct: 58.9 (#254)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Instruction Following12411131

Long Context GPT-4 leads

GPT-4: 37.7 (#207), Granite 3.1 8b Instruct: 35.2 (#236)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Longer Query12441162

Writing & Preference Too close to call

GPT-4: 35.0 (#263), Granite 3.1 8b Instruct: 35.7 (#261)

BenchmarkGPT-4Granite 3.1 8b Instruct
LMArena Text12631150
LMArena Creative Writing12441129
LMArena Multi-Turn12571108
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than Granite 3.1 8b Instruct?

Granite 3.1 8b Instruct is the stronger model overall, scoring 31.8 to 28.9 on the Noometry Index.

Is GPT-4 or Granite 3.1 8b Instruct better for coding?

Granite 3.1 8b Instruct scores higher on coding benchmarks: 33.9 versus 31.2 in the Noometry coding category.

How many benchmarks do GPT-4 and Granite 3.1 8b Instruct share?

12 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Granite 3.1 8b Instruct has 13.

Related comparisons

Go deeper