Language modelOpen weights

Granite 4.2 8B

IBM

Released: August 25, 2026
Data date: October 3, 2026

For Granite 4.2 8B, IBM adds agent tasks in real coding, terminal, and search environments to foundational training. This block distinguishes 8B from Granite 4.2 3B. It trains multi-step tool use against the task’s actual outcome.

The dense 8-billion-parameter checkpoint sits between 3B and 30B. Its weights use Apache 2.0.

Specifications and access

Specification	Value and source
Parameters	8 billionSource
Model class	Language model for agentic workflowsSource
License	Apache 2.0Source
Access	Open weights for local deploymentSource

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

SWE-bench Multilingual (Show measurement, test conditions, and source)

Source value: 30.78
Score: 30.78
Metric: SWE-bench Multilingual reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

SWE-Pro (Show measurement, test conditions, and source)

Source value: 19.11
Score: 19.11
Metric: SWE-Pro reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

SWE-Verified (Show measurement, test conditions, and source)

Source value: 47.67
Score: 47.67
Metric: SWE-Verified reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

Terminal-Bench 2.1 (Show measurement, test conditions, and source)

Source value: 20.56
Score: 20.56
Metric: Terminal-Bench 2.1 reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

tau3-bench (Show measurement, test conditions, and source)

Source value: 58.06
Score: 58.06
Metric: tau3-bench reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

BFCL v4 (Show measurement, test conditions, and source)

Source value: 50.29
Score: 50.29
Metric: BFCL v4 reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

ProfBench (Show measurement, test conditions, and source)

Source value: 41.2
Score: 41.2
Metric: ProfBench reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

BirdBench (Show measurement, test conditions, and source)

Source value: 41.07
Score: 41.07
Metric: BirdBench reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

GDPval (Show measurement, test conditions, and source)

Source value: 1189
Score: 1189
Metric: GDPval reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

AIME25 (Show measurement, test conditions, and source)

Source value: 86.67
Score: 86.67
Metric: AIME25 reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

GPQA (Show measurement, test conditions, and source)

Source value: 64.14
Score: 64.14
Metric: GPQA reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

LCBench v6 (Show measurement, test conditions, and source)

Source value: 73.24
Score: 73.24
Metric: LCBench v6 reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. IBM official technical breakdown.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

Page 1 of 2

Sources and data date

Every statement links to its underlying documentation or leaderboard.

Type	Evidence and data date
Research status	Research date October 4, 2026. 1 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Additional source	IBM Granite 4.2 announcement (retrieved October 3, 2026) · IBM Granite 4.2 announcement · Editorial description reviewed October 4, 2026
Additional source	IBM Granite 4.2 technical report (retrieved October 4, 2026) · IBM Granite 4.2 technical report · Editorial description reviewed October 4, 2026