Compares Grok Imagine against Nano Banana Pro/2.1. The prompt provided is intended for Grok Imagine ('Upload selfie + #GrokImagine prompt'), although the user claims NB Pro did it well too.
Overview
Compares Grok Imagine against Nano Banana Pro/2.1. The prompt provided is intended for Grok Imagine ('Upload selfie + #GrokImagine prompt'), although the user claims NB Pro did it well too. However, strictly speaking, the extracted prompt block is labeled for Grok Imagine usage in the context of the tweet's instruction. Wait, looking closer: 'I didn't think "Grok Imagine" would manage this editorial prompt as well as Nano Banana Pro/2.1.' Then 'Female fire-breather prompt below... Upload selfie + #GrokImagine prompt:'. The prompt itself is generic enough to apply to both, but the explicit labeling suggests it was tested on Grok. Given the ambiguity and the fact that it lists 'Nano Banana Pro/2.1', I will treat this as a valid prompt for Nano Banana Pro since the author asserts NB Pro handled it well. However, usually, if a prompt is explicitly tagged for another model in a comparison, we might skip. But here the prompt is shared as the one used for the comparison. Let's extract it as it is applicable to NB Pro per the author's claim.
The prompt
| 1 | Use the person reference only for facial identity and recognizable features; do not inherit its hair, wardrobe, pose or setting. |
| 2 | |
| 3 | Create a vertical late-night fashion-performance photograph inside a dark crowded club or cabaret, framed tightly from approximately the lower chest upward around one adult woman caught at the peak of a fire-breathing performance. |
| 4 | |
| 5 | Build the composition around a large real flame bursting diagonally and upward from her mouth. Tip her head back, extend the neck, keep the mouth open, and allow the flame to pass directly through the visual center of the face so it obscures part of the nose, eyes and forehead rather than floating neatly above her. The flame should be irregular, overbright and physically turbulent, with a white-hot center, yellow-orange edges, faint smoke and unstable shape. |
| 6 | |
| 7 | Stretch both arms outward toward opposite edges of the frame, elbows slightly bent and hands remaining active rather than posed decoratively. Let one hand control a dark performance torch or metal rod angled toward the mouth while the opposite hand stays extended with a second tool partly visible near the frame edge. Keep the shoulders broad and lifted beneath the action. |
| 8 | |
| 9 | Give her an oversized antique-gold sunburst headdress built from irregular radiating spikes and clustered metallic elements around the crown, wide enough to extend beyond both sides of the head. Do not make it perfectly symmetrical or pristine. Let individual points catch the firelight differently. |
| 10 | |
| 11 | Dress her in a dark bronze-gold beaded performance bodice with narrow straps, dense small metallic stones and irregular chain detailing. Layer several antique-gold and crystal necklaces closely around the neck and upper chest, with longer strands hanging unevenly below. Add dangling gold earrings and a few substantial metallic bracelets, but keep the headdress and flame dominant. |
| 12 | |
| 13 | Keep the hair mostly pulled back beneath the headpiece, with a few damp or loose dark strands around the temples and neck. Preserve strong eye makeup, defined brows and a deep red lip, but do not give the face clean beauty illumination. |
| 14 | |
| 15 | Let the visible flame be the dominant local light source. It should throw hot unstable amber light upward across the mouth, cheeks, chin, throat, jewelry and nearest headdress points while areas only slightly farther away drop abruptly into red-brown and near-black shadow. Allow portions of the flame to clip completely white. Add only weak dirty club ambience from the sides—small blue, violet or red contamination in the shadows—without creating a polished rim light or separate facial key. |
| 16 | |
| 17 | Keep several indistinct adult spectators visible behind her as soft, dark human shapes at different depths. Do not remove them or |
Similar prompts
See all →Anime celMake Japanese TV-anime style videos (MP4) with code — cel-shaded characters with tapered line art, big detailed eyes, hair clumps with a specular band, comedic anime reactions (sweat drop, anger vein, sparkle eyes, ≥< eyes, dot eyes, tears), painted skies with cumulus clouds, school/street/classroom/rooftop sets with time-of-day palettes, speed/focus lines, impact frames, dramatic zooms, smears, sakura petals and anime subtitles (CJK). Canvas 2D rendered frame by frame in headless Chrome, encoded with ffmpeg. Use when a reference video or brief calls for anime / 動畫 / 日系動畫 / アニメ / cel-shaded / slice-of-life / school romance / shōnen-style reactions, or when the video-clone router picks the anime-cel style. 2D only; characters are original (Hina, Ren) or new ones built on the same rig.
by edenfunf · ★ 3.6kLemo opuscarDirect and produce a short film made entirely in code, in one of the styles of the Lemo-Opuscar library (e.g. Impasto Oil Painting 油画厚涂, Watercolor Brush 水彩笔刷, Ukiyo-e 浮世绘, Satellite Briefing 卫星简报, Exploded Teardown 产品拆解, Microscope Story 显微叙事). Use when the user asks for a video, film, short, promo, explainer or animation (视频、短片、动画、宣传片) in a named style, including data, science and product explainers built from real material such as satellite imagery, laser scans, CAD files or archive scans (数据解说、科普、产品拆解); when they want a film made about their own topic and haven't chosen a style (help them choose); or when they ask which film styles there are. Not for editing or converting existing video files.
by lemomo-ai · ★ 3.0kXiaohei illustrationsGenerate Ian-style Xiaohei illustrations for articles, posts, blogs, Notion documents, workflows, methods, processes, structures, states, metaphors, and opinions. Use when the user asks for weird, hand-drawn, Xiaohei, article, editorial, body illustrations, illustration ideas, shot lists, title removal, image edits, or image regeneration. By default, all image text and annotations are English, even when the source article is Chinese; only use another language when the user explicitly asks.
by dzhng · ★ 1.3k
Scenario game assetsUse when creating game art through Scenario MCP: sprites, sprite sheets, icons, props, loot, tilesets and autotile terrain transitions, seamless tiles, isometric buildings, top-down maps, pixel art, UI buttons and panels, parallax background layers or depth planes for side-scrollers, and character or concept art; or when assets need transparent backgrounds, background removal, style-consistent variation batches, upscaling, pixel-grid cleanup, or PNG export for Unity, Godot, or Unreal.
by scenario-labs · ★ 854AI-Toolkit LoRA Trainer (WAN 2.2 & Z-Image)Train custom LoRAs with ostris AI-Toolkit. Covers WAN 2.2/2.1 (people, styles, video motion) and Z-Image (Turbo & Base, low-VRAM image LoRAs). Use when the user wants to train a WAN or Z-Image LoRA; covers local + RunPod setup, dataset prep, key params, and using the result in a ComfyUI workflow.
by artokun · ★ 759ANIMA 1.0 (Anima Base Ultra) Text-to-Image WorkflowsAnime/illustration text-to-image (ANIMA 1.0, ~2B Cosmos DiT). Use for anime, manga, illustrated characters; accepts Danbooru tags + natural language; runs/trains on <6GB VRAM; includes anime inpainting via Anima-LLLite ControlNetby artokun · ★ 759
by edenfunf · ★ 3.6kLemo opuscarDirect and produce a short film made entirely in code, in one of the styles of the Lemo-Opuscar library (e.g. Impasto Oil Painting 油画厚涂, Watercolor Brush 水彩笔刷, Ukiyo-e 浮世绘, Satellite Briefing 卫星简报, Exploded Teardown 产品拆解, Microscope Story 显微叙事). Use when the user asks for a video, film, short, promo, explainer or animation (视频、短片、动画、宣传片) in a named style, including data, science and product explainers built from real material such as satellite imagery, laser scans, CAD files or archive scans (数据解说、科普、产品拆解); when they want a film made about their own topic and haven't chosen a style (help them choose); or when they ask which film styles there are. Not for editing or converting existing video files.by lemomo-ai · ★ 3.0kXiaohei illustrationsGenerate Ian-style Xiaohei illustrations for articles, posts, blogs, Notion documents, workflows, methods, processes, structures, states, metaphors, and opinions. Use when the user asks for weird, hand-drawn, Xiaohei, article, editorial, body illustrations, illustration ideas, shot lists, title removal, image edits, or image regeneration. By default, all image text and annotations are English, even when the source article is Chinese; only use another language when the user explicitly asks.by dzhng · ★ 1.3kScenario game assetsUse when creating game art through Scenario MCP: sprites, sprite sheets, icons, props, loot, tilesets and autotile terrain transitions, seamless tiles, isometric buildings, top-down maps, pixel art, UI buttons and panels, parallax background layers or depth planes for side-scrollers, and character or concept art; or when assets need transparent backgrounds, background removal, style-consistent variation batches, upscaling, pixel-grid cleanup, or PNG export for Unity, Godot, or Unreal.by scenario-labs · ★ 854AI-Toolkit LoRA Trainer (WAN 2.2 & Z-Image)Train custom LoRAs with ostris AI-Toolkit. Covers WAN 2.2/2.1 (people, styles, video motion) and Z-Image (Turbo & Base, low-VRAM image LoRAs). Use when the user wants to train a WAN or Z-Image LoRA; covers local + RunPod setup, dataset prep, key params, and using the result in a ComfyUI workflow.by artokun · ★ 759ANIMA 1.0 (Anima Base Ultra) Text-to-Image WorkflowsAnime/illustration text-to-image (ANIMA 1.0, ~2B Cosmos DiT). Use for anime, manga, illustrated characters; accepts Danbooru tags + natural language; runs/trains on <6GB VRAM; includes anime inpainting via Anima-LLLite ControlNetby artokun · ★ 759Your agent, with every skill.
Connect once. Free when signed in.