Noometry

Model comparison

GPT-4 Turbo vs Granite 3.1 8b Instruct

Granite 3.1 8b Instruct is the stronger model overall, scoring 32.4 to 30.5 on the Noometry Index.

Last verified . 12 shared benchmarks.

GPT-4 Turbo OpenAI

30.5

Rank #292 Confirmed

Granite 3.1 8b Instruct IBM

32.4

Rank #258 Confirmed

Summary

  • They share 12 benchmarks with published results for both. GPT-4 Turbo scores higher in 4 categories and Granite 3.1 8b Instruct in 4 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where Granite 3.1 8b Instruct leads 33.0 to 9.0.
  • Granite 3.1 8b Instruct has downloadable open weights; the other is API-only.

Side by side

GPT-4 TurboGranite 3.1 8b Instruct
ProviderOpenAIIBM
Noometry Index30.532.4
Released2023-11-06—
WeightsProprietaryOpen
Context window128K—
Max output4K—
Input $ / M tokens$10—
Output $ / M tokens$30—
Results tracked3613

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

Category by category

Coding Too close to call

GPT-4 Turbo: 33.8 (#249), Granite 3.1 8b Instruct: 34.5 (#233)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Coding12681186
WeirdML18%—
BigCodeBench Instruct48.2%—
BigCodeBench Complete58.2%—
HumanEval+86.6%—
MBPP+73.3%—

Agentic & Tool Use Not comparable

GPT-4 Turbo: —, Granite 3.1 8b Instruct: 24.1 (#120)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
Berkeley Function Calling Leaderboard—27.1%
METR Time Horizons36.7%—

Reasoning Granite 3.1 8b Instruct leads

GPT-4 Turbo: 15.3 (#317), Granite 3.1 8b Instruct: 22.1 (#207)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Hard Prompts12511145
SimpleBench25.1%—
Chess Puzzles6%—
DTBench61.6%—
LMCA9.8%—
Epoch Capabilities Index127.25—
ForecastBench59.4—

Math Granite 3.1 8b Instruct leads

GPT-4 Turbo: 9.0 (#322), Granite 3.1 8b Instruct: 33.0 (#209)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Math12721152
FrontierMath (Tiers 1-3)0.7%—
OTIS Mock AIME 2024-20256.7%—
MATH Level 546.7%—

Knowledge Granite 3.1 8b Instruct leads

GPT-4 Turbo: 24.3 (#268), Granite 3.1 8b Instruct: 31.1 (#220)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Expert12231142
GPQA Diamond46.6%—
Confabulations28.4%—
MMLU81.3%—

Multimodal Not comparable

GPT-4 Turbo: 30.6 (#110), Granite 3.1 8b Instruct: —

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Vision1090—

Multilingual GPT-4 Turbo leads

GPT-4 Turbo: 40.5 (#216), Granite 3.1 8b Instruct: 30.9 (#260)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Non-English12451099
LMArena Chinese12421145
LMArena Russian12591092
LMArena French1276—
LMArena German1259—
LMArena Japanese1194—
LMArena Korean1187—
LMArena Spanish1260—

Instruction Following GPT-4 Turbo leads

GPT-4 Turbo: 65.8 (#216), Granite 3.1 8b Instruct: 58.6 (#259)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Instruction Following12491131

Long Context GPT-4 Turbo leads

GPT-4 Turbo: 38.0 (#206), Granite 3.1 8b Instruct: 35.2 (#241)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Longer Query12541162

Writing & Preference GPT-4 Turbo leads

GPT-4 Turbo: 47.7 (#206), Granite 3.1 8b Instruct: 35.5 (#266)

BenchmarkGPT-4 TurboGranite 3.1 8b Instruct
LMArena Text12721150
LMArena Creative Writing12691129
LMArena Multi-Turn12671108

Frequently asked questions

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

Granite 3.1 8b Instruct is the stronger model overall, scoring 32.4 to 30.5 on the Noometry Index.

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

They score almost the same on coding (33.8 vs 34.5); test both on your own repository before choosing.

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

12 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Granite 3.1 8b Instruct has 13.

Related comparisons

Go deeper