Granite 4.0 H Small

Last updated: Recently

Model Type

Text

Max Context

Max Output

Parameter Scale

Granite 4.0 H Small by IBM: Text model; TTFT 8.716s, 502.3 tok/s.

Input Modality

Output Modality

Inference Speed

502.349 tokens/s

Latest Release Date

9/22/2025

SDK Ecosystem

artificial-analysismanufactureraa-bootstrap

Model Overview Fields

name

Granite 4.0 H Small

Release date

9/22/2025

Performance

502.349 tokens/s

model_creator.name

IBM

AA Evaluation Scores

AA Intelligence Index

10.8

AA Coding Index

8.5

AA Math Index

13.7

MMLU Pro

0.624

GPQA

0.416

HLE

0.037

LiveCodeBench

0.251

SciCode

0.209

Math-500

AIME

AIME 2025

0.137

IFBench

0.315

LCR

0.09

TerminalBench Hard

0.023

TAU2

0.173

Output Speed (tokens/s)

502.349 tok/s

TTFT (s)

8.716s

First Answer Token (s)

8.716s

IBM

◎

Granite 4.0 H Small

$0.06/M$0.25/M

Granite 4.0 H Small by IBM: Text model; TTFT 8.716s, 502.3 tok/s.

There are no reviews for this model yet.

artificial-analysismanufactureraa-bootstrap

0.0