Nano Banana 2 Prompt Engineering for Headshots with Specific Clothing Changes
Creating professional headshots often requires versatility. A single photo might need to look appropriate for a corporate LinkedIn profile, a creative portfolio, or a casual social media post. While the Nano Banana 2 tool offers powerful text-to-image and image-to-image workflows, users frequently encounter challenges when trying to alter clothing without losing the subject's identity. It is crucial to understand that Nano Banana refers to the AI image generation and editing tool, not a skincare brand or physical product. When using this platform, prompt instructions describe desired outcomes but do not guarantee identity, label, object, or typography preservation.
The core use case here involves taking an existing portrait and instructing the model to swap a specific garment—such as changing a t-shirt to a blazer—while keeping the face, lighting, and background consistent. This workflow relies heavily on precise language and an understanding of the underlying model capabilities. Google documents Nano Banana 2 as Gemini 3.1 Flash Image (gemini-3.1-flash-image), which balances speed and quality. However, users must be aware that prompt engineering is an iterative process. There are no guarantees that the output will perfectly match the input identity every time, especially when complex changes like clothing swaps are requested.
Understanding Identity Preservation Limits
Before attempting to change clothes, it is vital to set realistic expectations regarding identity preservation. The documentation explicitly states that prompt instructions do not guarantee identity preservation. This means that even with highly detailed descriptions of the subject's face, the AI might subtly alter facial features, skin texture, or expression during the generation process. This limitation is inherent to the generative nature of the technology rather than a flaw in the user's prompting technique.
When working with Nano Banana 2, you are interacting with distinct Google image models. For instance, Nano Banana Pro utilizes Gemini 3 Pro Image, while Nano Banana 2 Lite uses Gemini 3.1 Flash Lite Image. It is important to note that Google describes Nano Banana 2 Lite as focused on speed and cost. It is not optimized for multiple reference inputs or multi-turn sequential editing. Therefore, if your goal is high-fidelity identity retention during complex clothing modifications, relying solely on the Lite version without understanding its constraints may yield inconsistent results. Always verify the specific model capabilities before starting a project that requires strict adherence to facial consistency.
Five Materially Different Usable Prompts
To help you navigate these challenges, here are five materially different prompt examples designed for clothing modification. These are labeled as examples because actual results vary based on the input image and model behavior. Each prompt targets a specific scenario and includes adjustments to maximize the chance of success.
Example 1: Formal Business Attire Swap
Use Case: Converting a casual photo into a professional corporate headshot. Prompt: "Keep the exact facial features, skin tone, and expression identical to the source image. Change the current casual t-shirt to a navy blue business suit jacket with a white collared shirt underneath. Maintain the original office background lighting." Adjustments: If the face looks slightly different, add more emphasis on "exact facial features" and reduce the complexity of the clothing description to focus the model on the face first.
Example 2: Creative Studio Look
Use Case: Transforming a standard portrait into an artistic, fashion-forward image. Prompt: "Preserve the subject's identity completely. Replace the current outfit with a vibrant, textured red velvet blazer and a black silk scarf. Keep the studio lighting soft and the background blurred. Do not change the hair style or makeup." Adjustments: If the colors bleed into the face, specify "keep the face neutral and unchanged" to isolate the clothing modification area.
Example 3: Seasonal Wardrobe Update
Use Case: Adapting a summer photo for a winter context. Prompt: "Maintain the same facial structure and eyes. Change the light summer dress to a heavy wool coat with a fur-lined hood. Ensure the snow effect in the background matches the new winter clothing. Keep the smile and pose exactly the same." Adjustments: If the background changes too much, explicitly state "do not alter the background" to force the model to focus only on the foreground subject.
Example 4: Uniform or Role-Specific Outfit
Use Case: Creating a themed headshot for a specific role or event. Prompt: "Identity must remain intact. Swap the current casual wear for a crisp white medical scrub top with a stethoscope around the neck. Keep the lighting bright and clinical. Do not change the facial expression or hair." Adjustments: If the stethoscope appears distorted, simplify the request to just the scrubs first, then add accessories in a subsequent turn if the model supports multi-turn editing (note: Nano Banana 2 Lite is not optimized for this).
Example 5: Minimalist Style Shift
Use Case: Simplifying a busy outfit for a clean aesthetic. Prompt: "Keep the face and hair exactly as they are. Change the patterned button-down shirt to a plain, solid black crew-neck sweater. Remove all jewelry and accessories. Maintain the original lighting conditions." Adjustments: If the sweater texture looks artificial, add keywords like "high resolution fabric texture" to guide the model toward realism.
Optimizing Your Workflow for Best Results
Success with Nano Banana 2 often depends on the iterative nature of prompting. Since there are no guarantees of identity preservation, you may need to run multiple variations of a prompt to find the one that best retains the subject's likeness. Start with simple clothing changes before attempting complex layering or accessory additions. Remember that the website has a Nano Banana 2 product page at /nanobanana2 where you can access the generator and explore the prompt library for additional inspiration.
For users requiring higher fidelity, consider exploring the Nano Banana Pro options, which leverage Gemini 3 Pro Image. However, always remember that Google model names and capabilities must not be presented as proof of identical features on this website without verification. The key is to treat the prompt as a strong suggestion rather than a command. By combining clear, descriptive language with an understanding of the tool's limitations, you can achieve impressive results in modifying headshot attire.
Note: All prompt examples provided above are illustrative. Actual outputs depend on the specific input image and the dynamic behavior of the AI model.