stealthmodels.

MODEL FIELD GUIDE

Qwen3.8 27B: Ground Truth

Apache-2.0 multimodal model with 27B parameters and Q4 weight files around 14–18 GB, within reach of a high-memory workstation.

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

A good fit for

Local coding assistants and multimodal agents where control over the weights matters and a smaller model leaves too much capability behind.

Context window
1,000,000tokens[1] [2]
Active parameters
27BparametersEstimated[1]
License
Apache-2.0[1]
Model factsArchitecture, context & capabilities
Developer
Qwen[1] [2]
Origin
China (Alibaba Group/Qwen organization context)[3]
Family
Qwen 3.8[1] [2]
Context window
1,000,000 tokens[1] [2]
Maximum output
131,072 tokens[2]
Architecture
Dense causal language model with vision encoder; hybrid Gated DeltaNet and gated attention; MTP-trained[1]
Total parameters
27B[1]
Active parameters per token
27BEstimated[1]
Input
text, image, video[1] [2]
Output
text[1] [2]
API access & pricingAlibaba Cloud Model Studio · $0.424 input / $1.696 output · per 1M tokens

Alibaba Cloud Model Studio; input 0.424 USD/1M; cached read 0.085/1M; output 1.696/1M; Representative documented regional rate; cache and region-specific rates are separate.

[2]
  • Alibaba Cloud Model StudioObserved 2026-10-06
    Input $0.424Output $1.696Cached read $0.085Cached write $0.53

    per 1M tokens

    Available

    qwen3.8-27b[2]
Run it locallyWeights, memory & deployment

single high-memory workstation GPU

[1] [4]
  • Unsloth Q8_029 GB storage · 8-bit[4]
  • Unsloth BF1654.7 GB storage · BF16[4]
  • MTP Q4_0 component1.37 GB storage · 4-bit[4]
Measurements & sources5 primary references · Speed & serving conditions

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

Frontier Models blindly judge the AI generation unbiased through an intricate process. Our researchers control for mistakes.

Overall score

Overall combines weighted task scores on a scale from 0 to 100. Repeating a task does not increase its weight. Estimated means the total includes an estimate for one missing scene.

Benchmaxxing

Benchmaxx flags models that score unusually well on the familiar Pelican task compared with their other benchmark results. The percentage measures that imbalance, not the probability of training contamination. Flagged Pelican scores are excluded from Overall. Pelican never counts toward General or Visual.

We keep some prompts private to discourage test-specific optimization.

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.

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.