stealthmodels.

MODEL FIELD GUIDE

DeepSeek V4.1 Flash: Pelican

Pairs a sparse reasoning backbone with 196B parameters of token-indexed Engram memory. Its encoder-decoder design sharply reduces the KV cache needed for long context.

Drawn by this model01 / 01
Pelican benchmark result generated by this model
SVG BENCHMARKPelicanView result

A good fit for

Agents that repeatedly read large documents, codebases or conversation histories. Engram host-memory offload is useful when GPU memory is the constraint.

Context window
1,000,000tokens[1] [14]
Decode active
16Bparameters[3] [4]
License
MIT[1]
Model factsArchitecture, context & capabilities
Developer
DeepSeek-AI[1]
Origin
China[2]
Family
DeepSeek V4.1[1]
Context window
1,000,000 tokens[1] [14]
License
MIT[1]
Architecture
20-layer encoder + 20-layer decoder with CSA2, Single-Pass mHC, Engram n-gram memory and DSpark/MTP[3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
Total parameters
763.21B[6] [8]
Prefill active parameters
8B[3] [4]
Decode active parameters
16B[3] [4]
Backbone parameters
551.57B[8] [3]
N-gram table parameters
196.61B[8] [5]
Input
text, image[1]
Output
text[1]
API access & pricingOpenRouter · $0.3 input / $1.2 output · per 1M tokens

Alibaba Cloud Model Studio; input 0.283 USD/1M; cached read 0.03/1M; output 1.131/1M; Representative busy-period rate; documented idle-period rates are lower and region-specific.

[16] [14]
  • OpenRouterObserved 2026-10-06
    Input $0.3Output $1.2Cached read $0.006

    per 1M tokens

    deepseek/deepseek-v4.1-flash[15]
  • Alibaba Cloud Model StudioObserved 2026-10-06
    Input $0.283Output $1.131Cached read $0.03

    per 1M tokens

    Available

    deepseek-v4.1-flash[16] [14]
Run it locallyWeights, memory & deployment

large multi-GPU / rack-scale

[1] [14] [17]
  • Exact released checkpoint tensor payload510.29 GB storage · native mixed released precision[8]

N-gram memory placement: Host RAM offload with asynchronous prefetch is supported; GPU residency depends on the serving engine and configuration.

[4]
Engram native storage
202.76 GB[8]
Remaining native storage
307.53 GB[8]

An independent NVMe implementation also exists; its reported performance has not been reproduced by us.

[12] [13]

The released Engram tables occupy about 203GB including scales. Other checkpoint tensors occupy about 308GB, before KV cache and runtime overhead. These are storage footprints, not minimum VRAM requirements.

[8] [4]
Measurements & sources17 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.