Noometry

Model comparison

Granite 4.0 Micro vs Mixtral 8x7B

Granite 4.0 Micro is the stronger model overall, scoring 29.0 to 27.1 on the Noometry Index.

Last verified . 6 shared benchmarks.

Granite 4.0 Micro IBM

29.0

Rank #318 Confirmed

Mixtral 8x7B Mistral AI

27.1

Rank #334 Confirmed

Summary

  • They share 6 benchmarks with published results for both. Granite 4.0 Micro scores higher in 3 categories and Mixtral 8x7B in 2 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in instruction following, where Granite 4.0 Micro leads 69.9 to 51.0.
  • The biggest single-benchmark swing is IFEval: 84.9% for Granite 4.0 Micro and 57.5% for Mixtral 8x7B.
  • Granite 4.0 Micro is cheaper at $0.017 / $0.11 per million input/output tokens, against $0.70 / $0.70 for Mixtral 8x7B.
  • Granite 4.0 Micro accepts more context: 131K tokens versus 32K.

Side by side

Granite 4.0 MicroMixtral 8x7B
ProviderIBMMistral AI
Noometry Index29.027.1
Released2025-10-022023-12-11
WeightsOpenOpen
Context window131K32K
Max output118K32K
Input $ / M tokens$0.017$0.70
Output $ / M tokens$0.11$0.70
Results tracked838

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

Category by category

Coding Not comparable

Granite 4.0 Micro: —, Mixtral 8x7B: 32.8 (#269)

BenchmarkGranite 4.0 MicroMixtral 8x7B
LMArena Coding—1126
HumanEval+—39.6%
MBPP+—49.7%

Reasoning Granite 4.0 Micro leads

Granite 4.0 Micro: 19.2 (#265), Mixtral 8x7B: 18.2 (#285)

BenchmarkGranite 4.0 MicroMixtral 8x7B
Chess Puzzles0%—
LMArena Hard Prompts—1115
DTBench—49.6%
Adversarial NLI—55.2%
Epoch Capabilities Index—118.47
ForecastBench—56.3
HellaSwag—86.7%
PIQA—83.6%
WinoGrande—77.2%

Math Mixtral 8x7B leads

Granite 4.0 Micro: 12.0 (#307), Mixtral 8x7B: 18.8 (#289)

BenchmarkGranite 4.0 MicroMixtral 8x7B
Omni-MATH20.9%10.5%
OTIS Mock AIME 2024-20252.8%—
LMArena Math—1147
MATH Level 5—10%
GSM8K—74.4%

Knowledge Mixtral 8x7B leads

Granite 4.0 Micro: 9.9 (#304), Mixtral 8x7B: 11.0 (#301)

BenchmarkGranite 4.0 MicroMixtral 8x7B
GPQA Diamond28.3%30.6%
MMLU-Pro39.5%33.5%
GPQA (HELM)30.7%29.6%
LMArena Expert—1088
ARC (AI2) Challenge—87.3%
MMLU—70.6%
OpenBookQA—85.8%
TriviaQA—82.2%

Multilingual Not comparable

Granite 4.0 Micro: —, Mixtral 8x7B: 29.6 (#266)

BenchmarkGranite 4.0 MicroMixtral 8x7B
LMArena Non-English—1077
LMArena Chinese—1055
LMArena French—1166
LMArena German—1114
LMArena Japanese—931
LMArena Korean—968
LMArena Russian—1090
LMArena Spanish—1111

Instruction Following Granite 4.0 Micro leads

Granite 4.0 Micro: 69.9 (#169), Mixtral 8x7B: 51.0 (#297)

BenchmarkGranite 4.0 MicroMixtral 8x7B
IFEval84.9%57.5%
LMArena Instruction Following—1109

Long Context Not comparable

Granite 4.0 Micro: —, Mixtral 8x7B: 33.4 (#260)

BenchmarkGranite 4.0 MicroMixtral 8x7B
LMArena Longer Query—1103

Writing & Preference Granite 4.0 Micro leads

Granite 4.0 Micro: 46.7 (#216), Mixtral 8x7B: 34.2 (#270)

BenchmarkGranite 4.0 MicroMixtral 8x7B
WildBench67%67.3%
LMArena Text—1132
LMArena Creative Writing—1109
LMArena Multi-Turn—1115

Frequently asked questions

Is Granite 4.0 Micro better than Mixtral 8x7B?

Granite 4.0 Micro is the stronger model overall, scoring 29.0 to 27.1 on the Noometry Index.

Which is cheaper, Granite 4.0 Micro or Mixtral 8x7B?

Granite 4.0 Micro is cheaper. It lists at $0.017 per million input tokens and $0.11 per million output tokens; Mixtral 8x7B lists at $0.70 and $0.70.

Which has the bigger context window?

Granite 4.0 Micro does, with 131K tokens against 32K.

How many benchmarks do Granite 4.0 Micro and Mixtral 8x7B share?

6 benchmarks have published results for both models. Granite 4.0 Micro has 8 scored results on Noometry and Mixtral 8x7B has 38.

Related comparisons

Go deeper