stealthmodels.

MODEL FIELD GUIDE

Gemma 4 26B A4B: Ground Truth

Google DeepMind’s open-weight mixture-of-experts model activates 3.8B parameters per token, with configurable thinking, image understanding and native function calling.

Drawn by this model01 / 01
Ground Truth benchmark result generated by this model
SVG BENCHMARKGround TruthView result

A good fit for

Self-hosted coding assistants and multimodal agents where inference efficiency and open weights matter.

Context window
262,144tokens[5]
Active parameters
3.8Bparameters[1]
License
Apache 2.0[1] [2]
Model factsArchitecture, context & capabilities
First available
2026-04-02[3]
Developer
Google DeepMind[1] [2]
Origin
United StatesEstimated[3]
Family
Gemma 4[1] [2]
Context window
262,144 tokens[5]
Architecture
Mixture-of-Experts multimodal transformer with interleaved sliding-window and global attention[1] [4]
Total parameters
25.2B[1]
Active parameters per token
3.8B[1]
Input
text, image[1]
Output
text[1]
API access & pricingOpenRouter · $0.09 input / $0.3 output · per 1M tokens

OpenRouter: $0.09 input / $0.30 output per 1M tokens, observed 2026-10-09; provider rates vary.

[6]
  • OpenRouterObserved 2026-10-09
    Input $0.09Output $0.3

    per 1M tokens

    Representative listed paid provider rate; provider rates vary.

    google/gemma-4-26b-a4b-it[6]
  • OpenRouterObserved 2026-10-09
    Input $0Output $0

    per 1M tokens

    Free model route listed by the provider on the observation date.

    google/gemma-4-26b-a4b-it:free[7]
Run it locallyWeights, memory & deployment
Measurements & sources7 primary references

StealthMark score

StealthMark measures fluid and visual intelligence of AI models.

General and Visual

General Intelligence

Perspective, anatomy, correct shadows, purposeful detail and physical consistency.

Visual Intelligence

Likeness, anatomy, detail, light, materials, composition and artistic expression.

Judging

Scores reflect how well each result meets the benchmark requirements.

Overall score

Overall combines weighted task scores on a scale from 0 to 100. Estimated means the total includes an estimate for one missing scene.

Scoring categories

Prompt & likeness

Subjects, actions and likeness.

Anatomy & construction

Coherent bodies, joints and machinery.

Space & placement

Perspective, scale and contact.

Light & reflections

Lighting, shadows and reflections.

Purposeful detail

Clear, purposeful small features.

Artistry & character

Composition, expression and drawing skill, including complexity and its execution.

Materials & effects

Surfaces, texture and movement.

Visual integrity

Clean shapes, layers and edges.

BENCHMARK RESULTS

Compare results

This link opens the same results in the same order.

StealthMark benchmark results

Reset benchmark preferences?

Return to the default view, clear comparison selections and restore the advertisement.