IBMAuto-detected

granite-4.0-micro

Name: granite-4.0-micro benchmark results
Creator: BenchLeader

Best configuration ranks #594 of 610 on the BenchLeader Index at 35.4 ±3.1. Last measured 2 Oct 2025stale: no new result in six months. Released 2 Oct 2025.

Blended price: –
Output speed: –
First answer: –
Context: 128k

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Overall index35
Reasoning24
Agents & tools35
Maths25
Knowledge27
Instruction following29
Long context28
Composite35

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmark	default	Source
GPQA Diamond	28.3%#254	Epoch AI Benchmarking Hub
GPQA Diamond (AA)not in index	33.6%#473	Artificial Analysis
Humanity's Last Exam (AA)not in index	5.0%#377	Artificial Analysis

Agents & tools

Benchmark	default	Source
Terminal-Bench Hard	1.5%#321	Artificial Analysis
τ²-Bench Telecom (AA)not in index	12.6%#358	Artificial Analysis

Maths

Benchmark	default	Source
OTIS Mock AIME	2.8%#234	Epoch AI Benchmarking Hub

Knowledge

Benchmark	default	Source
AA-Omniscience	-77.5#455	Artificial Analysis

Instruction following

Benchmark	default	Source
IFBench	24.8%#374	Artificial Analysis

Long context

Benchmark	default	Source
AA-LCR	7.0%#404	Artificial Analysis

Composite

Benchmark	default	Source
AA Intelligence Index	5.1#514	Artificial Analysis