Benchmark suiteLatest source data: Jul 14, 2026Checked: July 16, 2026

IBM benchmarks.

A lab-level profile for IBM, covering tracked models, category leaders, open/proprietary mix, speed, value, source coverage, and portfolio depth.

All labs

Lab rank

#27

By available Arena average

Models tracked

6

6 open-weight models

Portfolio coverage

19%

Average matched benchmark signals

Fastest model

0.43s

Granite 4.1 8b

Lab profile

Portfolio shape, leaders, and model coverage.

The lab page now starts with the portfolio read, then drills into category leaders, value rankings, and model-level coverage rather than repeating source panels.

Portfolio read

IBM has 6 tracked benchmark profiles.

Lab #27

The portfolio combines 0 proprietary models and 6 open-weight models. Its average available Arena rank is 263.9, with a best category rank of #92. Average profile coverage is 19%, so model-level gaps remain visible in the cards below.

Best Arena model

Granite 4.1 8b

Best category rank #92

Highest coverage

Granite 4.1 8b

56% matched signals

Fastest response

Granite 4.1 8b

0.43s first token

Access mix

Open versus proprietary

Proprietary

0 / 0%

Open weights

6 / 100%

Value leader

Granite 4.1 8b

Category leaders

Best model by benchmark lens.

Each card shows the highest-ranked model from this lab in a composite benchmark suite, with the global rank and index left intact.

Value

Quality, price, latency, and speed

Granite 4.1 8b

#11161

value index

Coding

Code Arena, SWE-bench, and output speed

Granite 4.1 8b

#10018

coding index

Agents

Search, document, code, and latency

Granite 4.1 8b

#12654

agent proxy

Reasoning

Text, document, vision, and mode signals

Granite 4.1 8b

#22262

reasoning proxy

Context

Document, search, and text proxy

Granite 4.1 8b

#23162

context proxy

IBM value ranking table

Models from this lab with both quality and commercial price matches.

Top 1 of 1

#111
Granite 4.1 8b

granite-4.1-8b

IBMOpen weights
61value index
Price
$0.063blended / 1M tokens
Latency
0.43stime to first token
Speed
111 tok/soutput speed
Quality
43quality signal
ArenaArtificial Analysis

Model portfolio

IBM model coverage map.

Every tracked model keeps its best Arena rank, coverage score, commercial metrics, and suite placement visible in one scannable grid.

Granite 4.1 8b

granite-4.1-8b

Open weights

Coverage

56%

Best Arena

#92

Code

Value

#111

value index

Coding

#100

coding index

Latency

0.43s

111 tok/s

$0.063

Ibm Granite H Small

ibm-granite-h-small

Open weights

Coverage

11%

Best Arena

#254

Text

Value

Not listed

No price match

Coding

Not listed

No coding index

Latency

Not listed

Not listed

Not listed

Granite 3.1 8b Instruct

granite-3.1-8b-instruct

Open weights

Coverage

11%

Best Arena

#302

Text

Value

Not listed

No price match

Coding

Not listed

No coding index

Latency

Not listed

Not listed

Not listed

Granite 3.0 8b Instruct

granite-3.0-8b-instruct

Open weights

Coverage

11%

Best Arena

#314

Text

Value

Not listed

No price match

Coding

Not listed

No coding index

Latency

Not listed

Not listed

Not listed

Granite 3.1 2b Instruct

granite-3.1-2b-instruct

Open weights

Coverage

11%

Best Arena

#319

Text

Value

Not listed

No price match

Coding

Not listed

No coding index

Latency

Not listed

Not listed

Not listed

Granite 3.0 2b Instruct

granite-3.0-2b-instruct

Open weights

Coverage

11%

Best Arena

#329

Text

Value

Not listed

No price match

Coding

Not listed

No coding index

Latency

Not listed

Not listed

Not listed