Noometry

Model comparison

GPT-4.1 mini vs Granite 4.2 3b

Granite 4.2 3b is the stronger model overall, scoring 39.0 to 33.5 on the Noometry Index.

Last verified . 11 shared benchmarks.

GPT-4.1 mini OpenAI

33.5

Rank #232 Confirmed

Granite 4.2 3b IBM

39.0

Rank #163 Confirmed

Summary

  • They share 11 benchmarks with published results for both. GPT-4.1 mini scores higher in 3 categories and Granite 4.2 3b in 4 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Granite 4.2 3b leads 25.0 to 10.9.
  • Granite 4.2 3b has downloadable open weights; the other is API-only.

Side by side

GPT-4.1 miniGranite 4.2 3b
ProviderOpenAIIBM
Noometry Index33.539.0
Released2025-04-14—
WeightsProprietaryOpen
Context window1.05M—
Max output33K—
Input $ / M tokens$0.40—
Output $ / M tokens$1.60—
Results tracked4711

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

Category by category

Coding Granite 4.2 3b leads

GPT-4.1 mini: 30.4 (#286), Granite 4.2 3b: 39.4 (#146)

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Coding13671363
SWE-bench Verified (bash only)23.9%—
Aider Polyglot32.4%—
SciCode40.4%—
WeirdML37.6%—
BigCodeBench Instruct48.9%—
CadEval16%—

Agentic & Tool Use Not comparable

GPT-4.1 mini: 33.0 (#52), Granite 4.2 3b: —

BenchmarkGPT-4.1 miniGranite 4.2 3b
Berkeley Function Calling Leaderboard50.5%—

Reasoning Granite 4.2 3b leads

GPT-4.1 mini: 10.9 (#335), Granite 4.2 3b: 25.0 (#136)

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Hard Prompts13491308
ARC-AGI-20%—
Kagi LLM Benchmark48.6%—
ARC-AGI-13.5%—
CritPt0%—
Chess Puzzles7%—
Mystery Game Puzzles7%—
DTBench68.8%—
LMCA21.1%—
Epoch Capabilities Index135.01—

Math Not comparable

GPT-4.1 mini: 23.1 (#264), Granite 4.2 3b: —

BenchmarkGPT-4.1 miniGranite 4.2 3b
FrontierMath (Tiers 1-3)6.7%—
OTIS Mock AIME 2024-202544.7%—
Omni-MATH49.1%—
LMArena Math1343—
MATH Level 587.3%—
FrontierMath (Feb 2025 set)4.5%—

Knowledge Too close to call

GPT-4.1 mini: 34.7 (#183), Granite 4.2 3b: 35.6 (#169)

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Expert13381313
GPQA Diamond65.8%—
SimpleQA Verified12.7%—
MMLU-Pro78.3%—
GPQA (HELM)61.4%—

Multimodal Not comparable

GPT-4.1 mini: 35.2 (#77), Granite 4.2 3b: —

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Vision1181—

Multilingual GPT-4.1 mini leads

GPT-4.1 mini: 45.9 (#161), Granite 4.2 3b: 42.4 (#193)

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Non-English13181268
LMArena Chinese13291268
LMArena Russian13241249
LMArena French1357—
LMArena German1349—
LMArena Japanese1291—
LMArena Korean1298—
LMArena Spanish1319—

Instruction Following GPT-4.1 mini leads

GPT-4.1 mini: 73.9 (#113), Granite 4.2 3b: 67.2 (#195)

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Instruction Following13321271
IFEval90.4%—

Long Context Granite 4.2 3b leads

GPT-4.1 mini: 31.8 (#270), Granite 4.2 3b: 39.3 (#179)

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Longer Query13441293
Fiction.LiveBench44.4%—

Writing & Preference Too close to call

GPT-4.1 mini: 48.8 (#193), Granite 4.2 3b: 47.9 (#202)

BenchmarkGPT-4.1 miniGranite 4.2 3b
LMArena Text13411295
LMArena Creative Writing13001215
LMArena Multi-Turn13541293
EQ-Bench Creative Writing1147—
WildBench83.8%—

Frequently asked questions

Is GPT-4.1 mini better than Granite 4.2 3b?

Granite 4.2 3b is the stronger model overall, scoring 39.0 to 33.5 on the Noometry Index.

Is GPT-4.1 mini or Granite 4.2 3b better for coding?

Granite 4.2 3b scores higher on coding benchmarks: 39.4 versus 30.4 in the Noometry coding category.

How many benchmarks do GPT-4.1 mini and Granite 4.2 3b share?

11 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Granite 4.2 3b has 11.

Related comparisons

Go deeper