Skip to main content
Van Data Team
Current modelIBM

Granite 4.2 30B benchmarks, pricing and specs

Granite 4.2 30B is an IBM model released on August 25, 2026 that takes text input and produces text. Its price is not published and accepts up to 128,000 tokens of context. It scores 1340 Elo on Arena Text, #48 of 49 current models (Arena, measured October 8, 2026).

Context window
128K
Arena Text
1340 Elo · #48/49

Data verified October 13, 2026 · Dataset 2026-10-13

Share this

Key facts

Provider
IBM
Model family
Granite 4.2
Release date
August 25, 2026
Status
Current
Access
open weights
API model id
Not reported
Context window
128,000 tokens
Max output
Not reported
Input
text
Output
text
Task families
text
Declared capabilities
reasoning, coding

Granite 4.2 30B pricing

UnitPrice (USD)TierAppliesSource
1M input tokensNot reportedstandardCurrent
1M output tokensNot reportedstandardCurrent
1M input tokensNot reportedcachedCurrent

Independent benchmark results

Scores come from independent evaluators that ran the model themselves, pinned to a dated snapshot. Rank is among current models in the same snapshot. IBM's own launch figures are listed under strengths with the publisher named, because they are not measured the same way.

BenchmarkScoreRankMeasuredSource
Arena Text1340 Elo#48 of 49October 8, 2026Arena · 2026-10-12

Strengths, cautions and best uses

Strengths

  • A 30B dense reasoning model under Apache 2.0; IBM reports 57 on SWE-Bench Verified and 89.17 on AIME25
  • Thinking, non-thinking and low-effort modes, with reasoning-augmented tool calling
  • A native 128K context, extendable to 512K

Cautions

  • IBM says it is fine-tuned mainly on English instruction data, so other languages may be weaker
  • IBM publishes no per-token API price for Granite 4.2

Best for

  • Self-hosted enterprise agents that need a permissive licence

Alternatives to Granite 4.2 30B

The closest current textmodels by price. The cheapest model that scores as well or better is listed first; failing that, the cheapest one within 3% of the benchmark's scale.

Compare Granite 4.2 30B with

Granite 4.2 30B FAQ

What is the context window of Granite 4.2 30B?

Granite 4.2 30B accepts up to 128,000 tokens of context. That ranks #52 of 57 current text models by context length.

How does Granite 4.2 30B score on independent benchmarks?

Granite 4.2 30B scores 1340 Elo on Arena Text, #48 of 49 current models (Arena, measured October 8, 2026). These are independent evaluations, not IBM's own launch figures.

Is Granite 4.2 30B open source or API only?

Granite 4.2 30B is released as open weights that you can run yourself.

What are the best alternatives to Granite 4.2 30B?

The closest current alternatives by price and measured performance are GLM-5.3-Flash (135 points higher on Arena Text, 1475 vs 1340), Claude Fable 5.1 (161 points higher on Arena Text, 1501 vs 1340), GLM-5.3 (138 points higher on Arena Text, 1478 vs 1340) and Gemini 3.8 Flash (157 points higher on Arena Text, 1497 vs 1340).

Sources

Cite this page

Van Data Team. "Granite 4.2 30B Benchmarks, Pricing and Specs." AI Model Benchmark Dataset 2026-10-13. https://vandatateam.com/ai-model-benchmarks/granite-4-2-30b

Reuse the data: JSON · CSV · or explore every model in the AI model benchmark comparison.

Production evaluation

Deciding whether Granite 4.2 30B fits your workload?

Public benchmarks narrow the shortlist. We test the finalists on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.

  • Evaluation cases and acceptance thresholds tied to real work
  • Model routing, fallbacks and cost controls
Compare all modelsDiscuss your evaluation