<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Prompt Image Atlas</title>
    <description>The latest articles on DEV Community by Prompt Image Atlas (@promptimageatlas).</description>
    <link>https://dev.to/promptimageatlas</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4189659%2Fe11cc51d-142b-4dad-86ac-69055fd57af4.png</url>
      <title>DEV Community: Prompt Image Atlas</title>
      <link>https://dev.to/promptimageatlas</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/promptimageatlas"/>
    <language>en</language>
    <item>
      <title>ChatGPT portrait prompts: start with the right kind of request</title>
      <dc:creator>Prompt Image Atlas</dc:creator>
      <pubDate>Sat, 17 Oct 2026 19:29:17 +0000</pubDate>
      <link>https://dev.to/promptimageatlas/chatgpt-portrait-prompts-start-with-the-right-kind-of-request-3h3i</link>
      <guid>https://dev.to/promptimageatlas/chatgpt-portrait-prompts-start-with-the-right-kind-of-request-3h3i</guid>
      <description>&lt;p&gt;&lt;em&gt;Prompt Image Atlas editorial team&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Figbfanenipfrvaqhvwfp.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Figbfanenipfrvaqhvwfp.png" alt="Portrait task planning diagram" width="800" height="420"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Diagram/photo composite: Urbanphhotolab / Wikimedia Commons, CC BY-SA 4.0; cropped and combined. &lt;a href="https://commons.wikimedia.org/wiki/File:Waken_Mouthwash_-_Still_Life_Photography.jpg" rel="noopener noreferrer"&gt;https://commons.wikimedia.org/wiki/File:Waken_Mouthwash_-_Still_Life_Photography.jpg&lt;/a&gt; ; &lt;a href="https://creativecommons.org/licenses/by-sa/4.0/" rel="noopener noreferrer"&gt;https://creativecommons.org/licenses/by-sa/4.0/&lt;/a&gt; . Not a ChatGPT output.&lt;/p&gt;

&lt;p&gt;You ask for a professional headshot and get a polished stranger. The jacket looks right; the person doesn't. Before adding another camera term, decide whether you're inventing a portrait or editing a photograph of someone who already exists. Those are different briefs, even when both start with “make a portrait.”&lt;/p&gt;

&lt;p&gt;These ChatGPT portrait prompts separate those jobs. The templates below are editorial starting points, not tested outputs: we have not run these prompts in ChatGPT or compared them with Midjourney. The accompanying photograph is a lighting reference, not a ChatGPT result.&lt;/p&gt;

&lt;h2&gt;
  
  
  Choose generation or reference-photo editing
&lt;/h2&gt;

&lt;p&gt;For text-only generation, describe a fictional adult: expression, clothes, framing, background and light. There is no uploaded identity to preserve. Don't judge the result by whether it resembles a real person you never supplied.&lt;/p&gt;

&lt;p&gt;For reference-photo editing, upload a photograph you have permission to use, then say what should stay and what should change. Request recognizable facial features, expression and skin texture where appropriate, but inspect the result rather than assuming identity has been preserved. OpenAI's &lt;a href="https://help.openai.com/en/articles/8932459-creating-images-in-chatgpt" rel="noopener noreferrer"&gt;image creation guide&lt;/a&gt; describes creating images and editing an existing image; available controls can depend on your current experience.&lt;/p&gt;

&lt;p&gt;If you only want advice for a real photo shoot, say “write a photography brief; do not generate an image.” A shot list and an image are different deliverables.&lt;/p&gt;

&lt;h2&gt;
  
  
  Business headshot templates
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;Text-only generation:&lt;/strong&gt;&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Create a photographic headshot of a fictional adult architect wearing a navy jacket over a plain cream shirt. Relaxed closed-mouth smile, eye-level camera, head and shoulders, quiet gray background, soft window light from camera left. Natural skin texture. Leave space above the hair for a square crop. No text or logos.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;&lt;strong&gt;Reference-photo edit:&lt;/strong&gt;&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Edit the uploaded portrait for a business profile. Keep the person's recognizable facial features, age and expression. Replace the background with a quiet gray wall and the clothing with a navy jacket and cream shirt. Use soft light from the left. Avoid smoothing away natural skin texture. Keep the full head visible.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Check the eyes, hairline and jaw against the source. If the face changes, simplify the edit before asking for more styling.&lt;/p&gt;

&lt;h2&gt;
  
  
  Cinematic portrait templates
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;Text-only generation:&lt;/strong&gt;&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Create a cinematic photographic portrait of a fictional adult musician outside a small theater at dusk. Waist-up framing, looking slightly past the camera, dark wool coat. Warm light from the doorway on the left, cooler ambient light behind. Restrained contrast, visible detail in the shadow side of the face. No lettering.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;&lt;strong&gt;Reference-photo edit:&lt;/strong&gt;&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Using the uploaded photograph as the identity reference, place the person outside a theater at dusk in a dark wool coat. Preserve recognizable features and the original expression. Add warm doorway light from the left and a subdued cool background. Keep the face readable; do not add dramatic facial changes.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;“Cinematic” alone leaves many choices open. Name a light source and a scene before adding a mood label.&lt;/p&gt;

&lt;h2&gt;
  
  
  Natural-light portrait templates
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;Text-only generation:&lt;/strong&gt;&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Create a candid photographic portrait of a fictional adult gardener beside a greenhouse. Chest-up framing, gentle smile, simple linen shirt. Soft overcast daylight, muted green background, realistic skin texture, ordinary everyday atmosphere. No studio spotlight or beauty retouching.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;&lt;strong&gt;Reference-photo edit:&lt;/strong&gt;&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Edit the uploaded portrait to suggest soft overcast daylight beside a greenhouse. Preserve the person's recognizable features and clothing. Change only the background and lighting. Keep skin texture and a believable shadow direction. Do not add accessories.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;h2&gt;
  
  
  Revise one decision at a time
&lt;/h2&gt;

&lt;p&gt;Save the initial request, then change one instruction: framing, background or light. Compare the crop, likeness when relevant, and unwanted alterations. A useful chatgpt portrait prompt is a clear brief you can revise, not a promise of a perfect face.&lt;/p&gt;

&lt;p&gt;For more framing ideas, browse our &lt;a href="https://promptimageatlas.com/blog/realistic-portrait-prompts.html" rel="noopener noreferrer"&gt;portrait prompt guide&lt;/a&gt; and &lt;a href="https://promptimageatlas.com/blog/ai-image-prompt-composition-negative-space.html" rel="noopener noreferrer"&gt;composition guide&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Originally published by Prompt Image Atlas: &lt;a href="https://promptimageatlas.com/blog/chatgpt-portrait-prompts.html" rel="noopener noreferrer"&gt;https://promptimageatlas.com/blog/chatgpt-portrait-prompts.html&lt;/a&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>chatgpt</category>
      <category>llm</category>
    </item>
    <item>
      <title>Midjourney lighting prompts: give the light somewhere to come from</title>
      <dc:creator>Prompt Image Atlas</dc:creator>
      <pubDate>Sat, 17 Oct 2026 00:42:56 +0000</pubDate>
      <link>https://dev.to/promptimageatlas/midjourney-lighting-prompts-give-the-light-somewhere-to-come-from-41oe</link>
      <guid>https://dev.to/promptimageatlas/midjourney-lighting-prompts-give-the-light-somewhere-to-come-from-41oe</guid>
      <description>&lt;p&gt;Your portrait has the right coat, the right street and a face you like. Then you notice the light: bright cheeks, glowing hair, no obvious source. Adding “cinematic lighting” leaves you with another attractive image and the same unanswered question. Where is that light coming from?&lt;/p&gt;

&lt;p&gt;Start with a source you could point to. A window on the left gives you something concrete to describe and something concrete to inspect. These Midjourney lighting prompts are starting templates, not a tested comparison: we have not run the examples below in Midjourney, and the photograph above is a real photograph, not a generated result.&lt;/p&gt;

&lt;h2&gt;
  
  
  Write a plain base prompt
&lt;/h2&gt;

&lt;p&gt;Keep the scene modest while you work on lighting. Use this fictional subject:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Editorial photograph of an adult ceramic artist sitting at a wooden workbench, clay-stained blue apron, hands resting beside an unfinished bowl, eye-level waist-up composition, plain workshop wall.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;That gives the light a face, fabric and a rough surface to act on. Save the base text separately. When you change the lighting, leave the subject and composition alone so you can see what else changed accidentally.&lt;/p&gt;

&lt;p&gt;Midjourney’s &lt;a href="https://docs.midjourney.com/hc/en-us/articles/32023408776205-Prompt-Basics" rel="noopener noreferrer"&gt;Prompt Basics&lt;/a&gt; recommends clear descriptions and identifies lighting as a detail worth specifying. You do not need a paragraph of camera jargon to name a window.&lt;/p&gt;

&lt;h2&gt;
  
  
  Try soft window light first
&lt;/h2&gt;

&lt;p&gt;Append this to the base:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;A large window just outside the frame on the left, soft daylight across the artist’s face, gentle shadow on the right cheek, subdued background, neutral color.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Look at the shadow side of the face. Is it still readable? Does the bowl appear to receive light from the same direction? If the whole scene becomes flat, try “a darker right side of the workshop” rather than adding several competing lighting styles.&lt;/p&gt;

&lt;p&gt;Use this setup when you want a quiet working portrait. Avoid adding sunset, neon and a studio spotlight to the same prompt unless you actually want several sources.&lt;/p&gt;

&lt;h2&gt;
  
  
  Make a harder side-light version
&lt;/h2&gt;

&lt;p&gt;Replace the window-light sentence with:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;A small uncovered lamp outside the frame on the left, hard side light, a distinct nose shadow toward the right, visible clay texture, dark workshop background.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Here the useful check is the edge of the shadows. You are asking for a sharper transition than in the window version. A moodier expression or a darker apron alone does not show that the lighting instruction worked.&lt;/p&gt;

&lt;p&gt;If you dislike the face falling into darkness, ask for a faint reflected fill from the right. Keep that addition in your notes; it changes the setup you are comparing.&lt;/p&gt;

&lt;h2&gt;
  
  
  Separate rim light from face light
&lt;/h2&gt;

&lt;p&gt;For a third direction, try:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;A light behind the artist on the right creating a narrow bright edge along the shoulder and hair, soft dim light on the face, dark wall, restrained highlights.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Check the outline against the background. A glowing border around every object is worth rejecting if your intended look is photographic. Explain which shoulder should catch the light, then inspect the hands and bowl for distracting highlights.&lt;/p&gt;

&lt;h2&gt;
  
  
  Keep a useful comparison record
&lt;/h2&gt;

&lt;p&gt;Save the full prompt, model version, aspect ratio and settings with each output. A fixed seed can help organize comparisons, but it does not prove that lighting was the only visual change; see Midjourney’s &lt;a href="https://docs.midjourney.com/hc/en-us/articles/32604356340877-Seeds" rel="noopener noreferrer"&gt;Seeds guide&lt;/a&gt; before relying on it.&lt;/p&gt;

&lt;p&gt;Compare shadow direction, shadow softness and whether the background competes with the face. Keep the image that suits your layout, even if it follows the requested lighting imperfectly. For space around the subject, use our &lt;a href="https://promptimageatlas.com/blog/ai-image-prompt-composition-negative-space.html" rel="noopener noreferrer"&gt;composition guide&lt;/a&gt;; for a different subject, start with the &lt;a href="https://promptimageatlas.com/blog/realistic-portrait-prompts.html" rel="noopener noreferrer"&gt;portrait prompt guide&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Written by Prompt Image Atlas editorial team.&lt;/p&gt;

&lt;p&gt;Originally published at &lt;a href="https://promptimageatlas.com/blog/midjourney-lighting-prompts.html" rel="noopener noreferrer"&gt;Prompt Image Atlas&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>ai</category>
      <category>design</category>
      <category>photography</category>
    </item>
  </channel>
</rss>
