Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

See the free Radar Brief
BenchLM
Google logo
Model A
Gemma 4 31B

Google

60.61/100

Supported · Public rank #66

90% interval 44.2–77.0

Gemma 4 31B vs Granite 4.2 8B

Updated August 31, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

IBM logo
Model B
Granite 4.2 8B

IBM

46.88/100

Estimated · Public rank #162

90% interval 35.4–58.4

Decision reading

Gemma 4 31B has the higher public score estimate, 60.61 versus 46.88, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

2 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

  • Long documents

    Prompts that approach the documented context limit

    Gemma 4 31B

    Gemma 4 31B has the larger documented context window.

    Confidence: documented

Show secondary and unsupported calls

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Chat turn cost

    1K fresh input + 500 output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Not enough matched evidence

    The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. Granite 4.2 8B does not fit this workload in one request. Gemma 4 31B has no comparable published API token rate. Granite 4.2 8B has no comparable published API token rate.

    Confidence: listed-rates

  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

Shared results
2
Gemma 4 31B only
6
Granite 4.2 8B only
12
Like-for-like categories
0 / 8

1 category uses different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Knowledge

Directional only
Gemma 4 31B
52.9
Granite 4.2 8B
72.2
Weighted basis
3 vs 2 rows
Reading
Directional only

Agentic

Not comparable
Gemma 4 31B
Not measured
Granite 4.2 8B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Coding

Not comparable
Gemma 4 31B
41.6
Granite 4.2 8B
36.5
Weighted basis
1 vs 3 rows
Reading
Not comparable

Reasoning

Not comparable
Gemma 4 31B
Not measured
Granite 4.2 8B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Math

Not comparable
Gemma 4 31B
Not measured
Granite 4.2 8B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
Gemma 4 31B
Not measured
Granite 4.2 8B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
Gemma 4 31B
76.9
Granite 4.2 8B
Not measured
Weighted basis
1 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
Gemma 4 31B
Not measured
Granite 4.2 8B
79.3
Weighted basis
0 vs 1 rows
Reading
Not comparable

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

Gemma 4 31B
Self-hosted; infrastructure cost varies
Fits in one request
Granite 4.2 8B
Self-hosted; infrastructure cost varies
Fits in one request

Gemma 4 31B has no comparable published API token rate. Granite 4.2 8B has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Gemma 4 31B
Self-hosted; infrastructure cost varies
Fits in one request
Granite 4.2 8B
Self-hosted; infrastructure cost varies
Fits in one request

Gemma 4 31B has no comparable published API token rate. Granite 4.2 8B has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Gemma 4 31B
Self-hosted; infrastructure cost varies
Fits in one request
Cached-input rate unavailable
Granite 4.2 8B
Self-hosted; infrastructure cost varies
Does not fit in one request
Cached-input rate unavailable

Granite 4.2 8B does not fit this workload in one request. Gemma 4 31B has no comparable published API token rate. Granite 4.2 8B has no comparable published API token rate.

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Gemma 4 31B

No comparable hosted API rate

Granite 4.2 8B

No comparable hosted API rate

IBM Granite 4.2 8B model card

Provider availability

Gemma 4 31B

Generally Available · Gemini API, Google AI Studio, open weights

Google Gemma Gemini API guide

Granite 4.2 8B

Not sourced

Reasoning profile

Gemma 4 31B

Reasoning

Granite 4.2 8B

Reasoning

Weight access

Gemma 4 31B

Open Weight

Granite 4.2 8B

Open Weight

License

Gemma 4 31B

Open Weight

Granite 4.2 8B

Open Weight

Release date

Gemma 4 31B

2026-04-02

Granite 4.2 8B

2026-08-25

If you already use one of these models
Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
Gemma 4 31B has the higher public score estimate, 60.61 versus 46.88, but the 90% score intervals overlap.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Gemma 4 31B has the larger documented window (256K).

Run the same representative tasks against both endpoints before changing production traffic.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

Gemma 4 31B
API / mo$0
Self-host / mo$429
Break-even—
Granite 4.2 8B
API / mo$0
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Model the full break-even

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence20 rows

Agentic

  • Gert Labs

    Gemma 4 31B35.26%
    Source
    Granite 4.2 8B—

    Not directly comparable

  • Terminal-Bench 2.1

    Gemma 4 31B—
    Granite 4.2 8B20.6%
    Source

    Not directly comparable

  • τ³-bench results

    Gemma 4 31B—
    Granite 4.2 8B58.1%
    Source

    Not directly comparable

  • BFCL v4

    Gemma 4 31B—
    Granite 4.2 8B52.4%
    Source

    Not directly comparable

Coding

  • SWE-Rebench

    Gemma 4 31B41.6%
    Source
    Granite 4.2 8B—

    Not directly comparable

  • React Native Evals

    Gemma 4 31B75.2%
    Source
    Granite 4.2 8B—

    Not directly comparable

  • SWE-bench Verified

    Gemma 4 31B—
    Granite 4.2 8B47.7%
    Source

    Not directly comparable

  • SWE-bench Pro

    Gemma 4 31B—
    Granite 4.2 8B19.1%
    Source

    Not directly comparable

  • SWE Multilingual

    Gemma 4 31B—
    Granite 4.2 8B30.8%
    Source

    Not directly comparable

  • Terminal-Bench 2.1

    Gemma 4 31B—
    Granite 4.2 8B20.6%
    Source

    Not directly comparable

  • LiveCodeBench v6

    Gemma 4 31B—
    Granite 4.2 8B73.2%
    Source

    Not directly comparable

  • SciCode

    Gemma 4 31B—
    Granite 4.2 8B36.1%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Gemma 4 31B84.3%
    Source
    Granite 4.2 8B64.1%
    Source

    Gemma 4 31B leads this result

  • MMLU-Pro

    Gemma 4 31B85.2%
    Source
    Granite 4.2 8B74.0%
    Source

    Gemma 4 31B leads this result

  • HLE

    Gemma 4 31B26.5%
    Source
    Granite 4.2 8B—

    Not directly comparable

  • HLE w/o tools

    Gemma 4 31B19.5%
    Source
    Granite 4.2 8B—

    Not directly comparable

Math

  • AIME 2025

    Gemma 4 31B—
    Granite 4.2 8B86.7%
    Source

    Not directly comparable

  • HMMT Feb 2025

    Gemma 4 31B—
    Granite 4.2 8B78.3%
    Source

    Not directly comparable

Multimodal

  • MMMU-Pro

    Gemma 4 31B76.9%
    Source
    Granite 4.2 8B—

    Not directly comparable

Instruction following

  • IFBench

    Gemma 4 31B—
    Granite 4.2 8B79.3%
    Source

    Not directly comparable

Frequently asked questions

Which is better, Gemma 4 31B or Granite 4.2 8B?

Gemma 4 31B has the higher public score estimate, 60.61 versus 46.88, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, Gemma 4 31B or Granite 4.2 8B?

The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.

Which is better for agentic tasks, Gemma 4 31B or Granite 4.2 8B?

The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.

Which costs less, Gemma 4 31B or Granite 4.2 8B?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Gemma 4 31B or Granite 4.2 8B?

Gemma 4 31B has the larger documented context window: 256K, compared with 128K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 31, 2026

Watch Gemma 4 31B vs Granite 4.2 8B

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.