Model serving Cost Optimization Benchmarks

Granite 4.1

Dense language model family in 3B, 8B, and 30B sizes with instruction-tuned variants.

IBM's Granite 4.1 family is a set of open-weight language instruct models, including 30B, 8B, and 3B variants, released under the Apache 2.0 license.

Pricing

Artificial Analysis · Aug 12, 2026, 1:02 PM

Input / 1M

$0.05

Output / 1M

$0.10

Blended / 1M

$0.063

Output TPS

118

TTFT (s)

0.46

Model Intelligence

Arena ranking

Benchmarkable

Model level

family

GPQA

0.48

HLE

0.04

IFBench

0.44

LCR

0.24

TerminalBench Hard

0.02

TAU2

0.42

Recent stories

1 linked story

releasePRIMARY2026-04-29

IBM releases Granite 4.1 30B/8B/3B open models under Apache 2.0

IBM released Granite 4.1 as three open instruct models, with third parties quickly surfacing token-efficiency and deployment access. The update matters for teams evaluating smaller open models for agent workloads where output-token burn and openness both affect production cost.