Image2Studio
Examples

gpt-image-2

LLM pretraining data-mixture sankey

Social Media Post prompt for infographics, ready to edit and generate in Image2Studio.

Research Paper FiguresInfographicssocial-media-post
Generate with this prompt
Social Media Post prompt for infographics, ready to edit and generate in Image2Studio.
Social Media Post prompt for infographics, ready to edit and generate in Image2Studio.
Prompt

Landscape 16:9 sankey diagram of a pretraining data mixture, three stages with translucent colored ribbons. LEFT (8 source blocks, heights proportional to tokens): "Common Crawl (web) 540B" (muted navy, largest), "arXiv papers 180B" (dusty teal), "GitHub code 160B" (slate gray), "Wikipedia 40B" (soft terracotta), "StackExchange QA 30B" (warm copper), "Books (public domain) 25B" (pale olive), "Patents 18B" (pale navy), "Curated news & forums 15B" (dusty teal). MIDDLE (3 processing blocks, stacked): "Deduplicated (MinHash + exact)", "Quality-filtered (classifier + heuristics)", "PII-scrubbed (regex + NER)". RIGHT (3 final splits): "Pretraining set 1.4T tokens" (largest), "Instruction-tune pool 12B tokens", "RLHF preference pool 3B tokens". Flow ribbons inherit source color with mid-labels showing token counts ("85B", "320B", "44B"). Legend strip at bottom. Title: "LLM pretraining data mixture and downstream splits". Subtitle: "token counts after deduplication and quality filtering; ribbon thickness ∝ token flow."

Prompt structure
  • Modelgpt-image-2
  • Use casesocial-media-post
  • Styleprompt-defined style
  • Aspect ratiolandscape
Best used for
  • Use casesocial-media-post
  • Styleprompt-defined style
  • Recommended specslandscape
  • ModelGPT Image 2
How to customize

Use this LLM pretraining data-mixture sankey prompt as a base. Replace the subject, product details, environment, lighting, brand colors, and output ratio while keeping the social-media-post intent and the visual style direction stable.

Editable variables
  • SubjectReplace the product, person, scene, interface, or object being generated.
  • Use caseKeep or rewrite the destination as social media post, ad creative, cover image, poster, or ecommerce hero.
  • Framelandscape
  • Brand cuesAdd colors, materials, props, text-safe space, and the platform where the image will be used.
Recommended specs

Choose the frame for the destination first, then use 1K, 2K, or 4K based on output quality needs. Product images and posters usually benefit from higher resolution.

Pre-generation checklist
  • CheckSubject, lighting, composition, and frame are specific enough to review.
  • CheckText-safe space is included when the image will carry a headline or ad copy.
  • CheckResolution, credit cost, and final destination are checked before generation.
Why this prompt works

LLM pretraining data-mixture sankey works because it starts from the image job, then controls subject, composition, lighting, style, and frame. It is best for social media post while keeping the prompt-defined style direction stable.

Common questions
  • Can I use it directly?Yes. Replace the subject, brand cues, and frame, then generate in Image2Studio.
  • Which ratio should I use?landscape

Related prompts

Browse all prompts
Social Media Post prompt for infographics, ready to edit and generate in Image2Studio.
Research figures

Patient cohort and multimodal biomarker workflow

Create a Nature Medicine / Science Translational Medicine style research paper figure, landscape 3:2 (1536×1024), soft literature-science palette, minimal and elegant. Figure title: "Patient cohort and multimodal biomarker workflow". Layout: a clean 4-panel academic figure labeled A–D with small bold panel letters. A. CONSORT-style patient cohort flow diagram: "Screened n=1,248" → "Eligible n=612" → branch into "Training cohort n=428" and "External validation n=184". Include exclusion side boxes: "missing imaging n=81", "insufficient follow-up n=43", "quality-control fail n=32". B. Multimodal sample-processing flow: icons for "CT imaging", "blood proteomics", "EHR timeline", "outcome labels" flowing into a pale-blue fusion box "feature harmonization". C. Small Kaplan–Meier survival plot with two clean curves labeled "low-risk" and "high-risk", muted teal vs soft rose, x-axis "Months", y-axis "Event-free survival". D. Compact table-style performance summary with three rows: "AUROC", "C-index", "Calibration slope" and two columns "Internal" / "External". Style requirements: white background, light gray axes, thin lines, ample margins, muted teal, dusty blue, soft coral, pale sand, no neon, no dark background, Nature journal figure aesthetics, readable labels, precise arrows, subtle gridlines, no decorative clutter, no fake logos, no watermark.

Generate with this prompt
Brand Identity prompt for infographics, ready to edit and generate in Image2Studio.
Research figures

Single-cell immune atlas reveals treatment-response states

Create a polished Nature / Cell style biomedical research figure, landscape 3:2 (1536×1024), soft minimal palette, publication-ready. Figure title: "Single-cell immune atlas reveals treatment-response states". Layout: 4-panel figure labeled A–D. A. Large UMAP scatter plot with 8 softly colored immune clusters; labels: "CD8 T", "CD4 T", "B cells", "NK", "Mono", "DC", "Treg", "Plasma". Use pastel teal, sage, lavender, peach, slate, amber. B. Dot plot of marker genes with rows "GZMB", "IFNG", "CXCL13", "MS4A1", "LYZ", "FOXP3" and columns matching immune clusters; dot size = fraction, color = expression. C. Small stacked bar chart comparing "Responder" vs "Non-responder" cell-state proportions, with 5 muted segments and a tidy legend. D. Pseudotime trajectory diagram: a clean branching curve from "naive" to "effector" and "exhausted", with small arrows and gradient color. Style requirements: literature-science design, white background, thin gray axes, compact legends, readable micro-labels, restrained typography, soft colors, elegant spacing, no 3D, no glossy UI, no fake journal logo, no watermark.

Generate with this prompt
Social Media Post prompt for infographics, ready to edit and generate in Image2Studio.
Research figures

Multimodal medical-AI method figure

Create a Nature Biomedical Engineering / NeurIPS medical-AI method figure, landscape 3:2 (1536×1024), soft literature-science colors and minimal academic layout. Figure title: "Multimodal foundation model for clinical decision support". Layout: a left-to-right method pipeline with three horizontal bands and panel labels A–C. A. Inputs on the left: small clean icons and labeled cards "Radiology image", "Pathology tile", "EHR sequence", "Lab values", "Genomics". Use subtle rounded rectangles. B. Middle architecture: five modality encoders feeding into a central pale-teal block "Shared clinical representation"; include small modules "contrastive alignment", "missing-modality mask", "temporal attention". Add thin arrows and skip connections. C. Outputs on the right: three task heads "diagnosis", "risk score", "treatment response" with small calibrated probability bars. Add a lower inset "external validation" showing two hospital icons and an arrow labeled "site transfer". Style requirements: soft Nature/Science palette (muted teal, dusty blue, sage green, warm sand, coral accents), white background, precise vector-like arrows, modest shadows only, readable labels, lots of whitespace, no futuristic HUD, no clinical gore, no real hospital logos, no watermark.

Generate with this prompt
Social Media Post prompt for infographics, ready to edit and generate in Image2Studio.
Research figures

Therapeutic response bar and forest plot

Create a Nature Medicine style statistical results figure, landscape 3:2 (1536×1024), soft, restrained, publication-quality. Figure title: "Therapeutic response across molecular subgroups". Layout: 4-panel figure labeled A–D. A. Grouped bar chart: response rate (%) for four subgroups "A", "B", "C", "D" across two treatments "standard" and "adaptive". Use muted navy and soft teal bars, thin error bars, numeric labels. B. Forest plot of hazard ratios for subgroups with a vertical reference line at HR=1.0; rows "age <65", "age ≥65", "high inflammation", "low inflammation", "mutation-positive", "mutation-negative". Use small squares and confidence intervals. C. Volcano-style biomarker association plot with pale gray background points and highlighted labeled markers "IL6", "CXCL10", "TP53", "MKI67". D. Minimal mechanism schematic: adaptive therapy reduces inflammatory signaling and restores immune surveillance; use three clean nodes connected by arrows, no complex biology drawings. Style requirements: literature-science aesthetic, white background, soft desaturated colors, thin gray axes, clear legends, compact labels, generous margins, Nature-style figure polish, no fake values that look too random, no decorative background, no watermark.

Generate with this prompt
Social Media Post prompt for social ads, ready to edit and generate in Image2Studio.
Research figures

Transformer encoder–decoder architecture

Landscape 16:9 academic concept figure of the Transformer encoder-decoder architecture, NeurIPS camera-ready style. Two vertical column stacks side-by-side with a dashed divider. LEFT column header: "ENCODER (×N)". Blocks bottom-to-top: "Input tokens" → "Input Embedding" → "+ Positional Encoding" → dashed "Encoder layer" containing "Multi-Head Self-Attention", "Add & Norm", "Feed-Forward", "Add & Norm", with thin curved residual arrows around each sublayer. RIGHT column header: "DECODER (×N)". Blocks bottom-to-top: "Output tokens (shifted right)" → "Output Embedding" → "+ Positional Encoding" → dashed "Decoder layer" containing "Masked Multi-Head Self-Attention", "Add & Norm", "Multi-Head Cross-Attention" (horizontal arrow from encoder top labeled "keys, values"), "Add & Norm", "Feed-Forward", "Add & Norm". Above decoder: "Linear", "Softmax", "Output probabilities". Title: "Transformer: encoder–decoder with multi-head attention". Subtitle: "Vaswani et al., 2017".

Generate with this prompt
Social Media Post prompt for social ads, ready to edit and generate in Image2Studio.
Research figures

Denoising diffusion forward/reverse chain

Landscape 16:9 academic figure of diffusion forward + reverse chains, two horizontal chains stacked vertically. TOP chain (left→right) labeled "Forward diffusion q(x_t | x_{t-1})": five frames "x_0", "x_{T/4}", "x_{T/2}", "x_{3T/4}", "x_T" progressing from a crisp small mountain-sun landscape to pure Gaussian noise. Arrows between frames labeled "+ β_t ε". BOTTOM chain (right→left) labeled "Reverse denoising p_θ(x_{t-1} | x_t)": same five frames in reverse, with a small hexagonal ε_θ(x_t, t) block between each pair. Far-right curved arrow "T diffusion steps" connecting top-right to bottom-right; far-left curved arrow "sample x_0" connecting bottom-left to top-left. Title: "Denoising Diffusion: forward corruption and learned reverse". Subtitle: "Ho et al., 2020".

Generate with this prompt