Cohere — Enterprise AI

Platform
Websites
Category
AI
Style
Light, Minimal
Type
Cohere Text, Unica77 Cohere
Updated
Sep 2026

Build with AI

An original design in the taste of Cohere — structure, tokens, font swaps and an imagery plan (1,609 words).

Paste into any AI tool. Every screenshot in the prompt is a public link the tool can open, and you can attach the screenshot link as a visual reference. It tells the agent to build something original: new copy, your brand, licensed or generated images.

Open in

Long prompts are copied to your clipboard first — paste them in when the tool opens.

Connect via MCP

claude mcp add uidiscovery \
  --env UIDISCOVERY_URL=https://uidiscovery.com \
  -- node /path/to/uidiscovery/mcp/server.mjs

Then ask your agent (it calls get_design_brief({ slug: "cohere" }) and get_image_plan({ slug: "cohere" })):

Use the uidiscovery MCP server: call get_design_brief({ slug: "cohere" }) and get_image_plan({ slug: "cohere" }), look at the preview images, then build an original site for [my brand] that follows the brief. Cite https://uidiscovery.com/sites/cohere.

Download kit ↓Brief as Markdown ↗API docs
Preview the prompt
# Build brief — an original marketing website in the taste of Cohere

> Reference: https://cohere.com · UIDiscovery: https://uidiscovery.com/sites/cohere · captured 2026-09-21 at a 1440px desktop viewport (+ 390px iPhone shots) · [full page](https://uidiscovery.com/media/captures/cohere/full.webp) · [scroll video](https://uidiscovery.com/media/captures/cohere/preview.mp4)

## 1. Goal
Build an original marketing website for **[the user's brand]** that feels as considered as Cohere — same structure, rhythm and polish, entirely new identity and content.

## 2. Originality rules
- Build an ORIGINAL design for the user's own brand in the same *taste* as Cohere: structure, rhythm, hierarchy, palette mood and type pairing.
- Never reuse Cohere's logo, name, product imagery, illustrations, icons, screenshots or copy. Write new copy for the user's product.
- If the user hasn't said what their product is, ask first (or use clearly marked placeholders like [Brand], [Product], [Audience]).
- Reference screenshots are for looking at, not for embedding or tracing.

## 3. The taste, in short
- Cohere (cohere.com): AI marketing website, light, minimal.
- 2 sections: Hero → Call to action.
- Spacing: long, content-rich sections (median ≈1700px tall at 1440).
- Hero headline: 4 words — short and punchy; keep yours under ~9.
- Type: one family (Cohere Text). Imagery: mostly photography, some illustration.

## 4. Page structure (top to bottom)

### 1. Hero — state what the product is and for whom in one glance, and offer the primary action.
- Layout: headline-led stack: eyebrow → H1 → supporting line → actions, with a visual beside or below. Tall, ≈1.9 screens (1700px at 1440), own light band #ffffff; 65 image elements (icons, logos, product UI — not 65 photos); 10 buttons (tabs, toggles or per-card actions), 30 links (link lists / navigation).
- Slots: Eyebrow / announcement pill (optional); H1 (~4 words, like the reference); Supporting line, 1–2 sentences; Primary + secondary action; Hero visual (see image plan).
- Topic (tone/length only — never reuse): “Your AI. Your rules.”
- [Screenshot](https://uidiscovery.com/media/captures/cohere/s-01.webp) · imagery: slot 1

### 2. Call to action — close the page with one clear next step.
- Layout: short headline + one primary action on a contrasting band. Tall, ≈1.6 screens (1430px at 1440), own light band #ffffff; 11 images/icons (a set of cards or a small gallery); 11 buttons (tabs, toggles or per-card actions), 135 links (link lists / navigation).
- Slots: Headline (~6 words, like the reference); One sentence of support; Primary + secondary action.
- Topic (tone/length only — never reuse): “Ready to put AI to work?”
- [Screenshot](https://uidiscovery.com/media/captures/cohere/s-02.webp) · imagery: slot 2

## 5. Design system
### Colour
| Role | Hex | Use | Contrast on background |
|---|---|---|---|
| background | `#fafafa` | page background | — |
| foreground | `#000000` | body text, headings | 20.1:1 |
| accent | `#000000` | none captured — monochrome. Primary buttons invert (foreground fill, background text); add your brand colour only for small highlights | — |
| surface | `#ffffff` | cards, inputs, alternate bands | text on it 21.0:1 |
| muted | `#73738a` | secondary text, borders | 4.4:1 — borders / large text only |

Captured palette: `#fafafa` `#ffffff` `#000000` `#73738a` `#17171c` `#212121`. Roles are assigned automatically — sanity-check them against the screenshot. Keep the palette's *mood* (light, neutral, desaturated) and proportions; swap the accent for the user's brand colour.

### Tokens
```css
:root {
  --background: #fafafa;
  --foreground: #000000;
  --accent: #000000;
  --surface: #ffffff;
  --muted: #73738a;
  --font-heading: "Inter", ui-sans-serif, system-ui, sans-serif;
  --font-body: "Inter", ui-sans-serif, system-ui, sans-serif;
  --radius: 12px; /* calibrate from the screenshot */
}

/* Tailwind v4 — in globals.css after @import "tailwindcss"; */
@theme {
  --color-background: var(--background);
  --color-foreground: var(--foreground);
  --color-accent: var(--accent);
  --color-surface: var(--surface);
  --color-muted: var(--muted);
  --font-heading: var(--font-heading);
  --font-sans: var(--font-body);
}
```

### Typography
- Headings: **Inter** (Google Fonts) — reference uses Cohere Text. “Cohere Text” is likely a custom or licensed face. Inter is a neutral stand-in — pick a closer free match by eye from the screenshot.
- Body: same family — a single-family system; create contrast with size and weight, not a second face.
- Scale (starting point, sizes weren't measured — calibrate against the 1440px screenshot): display clamp(2.75rem, 6vw, 5rem), h2 clamp(1.75rem, 3vw, 2.5rem), h3 1.25rem, body 1rem–1.125rem / 1.6, small 0.875rem. Headings: tight tracking (-0.02em), line-height 1.05–1.15.

### Spacing & rhythm
Sections run ≈1700px tall at 1440 (median; range 1430–1700px, captures cap at 1700px) — long, content-rich chapters. Start with 96–128px vertical padding per section on desktop and about two-thirds of that on mobile. One continuous background — separate sections with space and type, not colour bands. 8px spacing grid, 1200–1280px max content width, 24px (mobile) / 32–48px (desktop) gutters.

### Shape, depth & theme
Flat surfaces, hairline 1px borders in the muted colour, small radii (4–8px), almost no shadow. Read exact corner radius, border and shadow treatment from the screenshots. Light-first; for a dark mode use a near-black (not #000) background, ~90% white text, lift surfaces with lightness not shadow, re-check the accent.

## 6. Imagery plan

Best guess: mostly **photography**, some illustration (light, neutral, desaturated, AI). Inferred from element counts, colours and industry — not from the pixels — so open the linked screenshots and match the real style before sourcing anything.

### Slot 1 · 1. Hero → photography
- Photography, seen in the capture (people). (measured from the screenshot) Count: 1–3 key visuals; the rest are icons/avatars. Aspect: 16:9 full-bleed (or 21:9). Look first: [ref 1](https://uidiscovery.com/media/captures/cohere/s-01.webp)
- **Stock** — “abstract gradient mesh”: [Unsplash](https://unsplash.com/s/photos/abstract-gradient-mesh?orientation=landscape) · [Pexels](https://www.pexels.com/search/abstract%20gradient%20mesh/?orientation=landscape) · [Pixabay](https://pixabay.com/images/search/abstract%20gradient%20mesh/?orientation=horizontal)
- **Generate** — `Photograph: abstract gradient mesh. Composition: wide shot, subject off-centre to one side, generous clean negative space for the headline. Lighting: soft natural daylight, airy and high-key. Neutral, desaturated colour grade that sits on #ffffff, #000000. Photorealistic, natural textures, shallow depth of field, 35mm lens. Aspect ratio 16:9. No text, no letters, no logos, no watermarks, no brand marks or trademarked products, no recognisable real people.`

### Slot 2 · 2. Call to action → illustration (uncertain)
- Light AI section with a few graphics — illustration, icon art or a product shot. Count: a set of ~11 in one consistent style. Aspect: 21:9 (background). Look first: [ref 1](https://uidiscovery.com/media/captures/cohere/s-02.webp)
- **Stock** — “technology flat illustration”: [Unsplash](https://unsplash.com/s/photos/technology-flat-illustration?orientation=landscape) · [Pexels](https://www.pexels.com/search/technology%20flat%20illustration/?orientation=landscape) · [Pixabay](https://pixabay.com/images/search/technology%20flat%20illustration/?orientation=horizontal)
- **Generate** — `Flat vector illustration: abstract gradient mesh, friendly simplified shapes, limited palette of #ffffff, #000000 plus one soft neutral, plenty of empty background, consistent line weight, neutral, desaturated mood. Aspect ratio 21:9. No text, no letters, no logos, no watermarks, no brand marks or trademarked products, no recognisable real people.`
- Uncertain: this could equally be photography, a product shot or icon art — open the reference screenshots and match what you see.


### If the screenshots show a different style
Open the refs first and name what you actually see (photo / product UI / illustration / abstract gradient or 3D / type only). If it differs from the guess, switch routes — UI is built in code, type-only needs no image, otherwise:
- **abstract 3D / gradient** — generate `Abstract 3D render: soft, matte clay shapes and a smooth gradient in #73738a and #ffffff on a #fafafa background. Lighting: soft natural daylight, airy and high-key; neutral, desaturated tone. Minimal and calm, generous negative space on one side for text. Aspect ratio 16:9. No text, no letters, no logos, no watermarks, no brand marks or trademarked products, no recognisable real people.` · stock: [“abstract gradient 3d” on Unsplash](https://unsplash.com/s/photos/abstract-gradient-3d?orientation=landscape) (same query works on Pexels/Pixabay)

- **Licences:** Unsplash and Pexels images are free for commercial use under their own licences; Pixabay under the Pixabay Content License. Check each image's page before use, credit the photographer when asked (good practice anyway), and skip images showing logos, trademarks, branded products or identifiable people without a release.
- **AI generation:** the prompts work with any image model (GPT Image, Imagen, Flux, Midjourney, Firefly, SDXL…) — check its commercial-use terms. Never put the reference brand's name or “in the style of <brand>” in a prompt.
- **Never** download, trace or re-host the reference's own images, illustrations, product shots or logos.

## 7. Build steps
1. Confirm brand, product, audience and the primary action with the user.
2. Set up the tokens (CSS variables / Tailwind @theme) and load the fonts.
3. Build the sections in order, mobile-first, with each reference screenshot open.
4. Write original copy for every slot — match length and tone, never the words.
5. Source every image via the imagery plan (stock or generated); compress it and record credits.
6. Run the checklist at 390 / 768 / 1440px and fix what fails.

## 8. Acceptance checklist
- [ ] Responsive at 390px, 768px and 1440px — no horizontal scroll, tap targets ≥ 44px on mobile.
- [ ] Colour contrast WCAG AA: 4.5:1 body text, 3:1 large text and UI; visible focus states; full keyboard navigation.
- [ ] Semantic HTML: one <h1>, ordered headings, landmarks (header/nav/main/footer), labelled form fields, alt text on meaningful images.
- [ ] Performance: modern image formats with width/height set, lazy-load below the fold, ≤ 2 font families with font-display: swap, LCP < 2.5s.
- [ ] prefers-reduced-motion respected; animations under ~300ms and never blocking content.
- [ ] Originality: no copied logos, images, illustrations or copy from the reference; every image licensed or generated; credits recorded.

## 9. If you use Next.js + Tailwind
Paste the @theme block into globals.css, load fonts with next/font/google, make each section a server component in components/sections/, use next/image with explicit sizes, and keep client components for interactive bits only (menus, toggles, accordions).

---
More: tokens https://uidiscovery.com/api/v1/sites/cohere/tokens · kit https://uidiscovery.com/api/v1/sites/cohere/kit · MCP `get_design_brief({ slug: "cohere" })`, `get_image_plan({ slug: "cohere" })`. Cite the reference as https://uidiscovery.com/sites/cohere.

Visit cohere.com ↗Download kit ↓

2 sections found. 0 selected for comparison.

About this design

Written from the capture data for Cohere.

Cohere (cohere.com) is a website in the AI category. Its homepage is captured as 2 sections, top to bottom: hero and call to action. The hero leads with “Your AI. Your rules.”.

It's a light design: the page background is #fafafa with #ffffff as the main text colour and #73738a as an accent. Headings are set in Cohere Text, body copy in Unica77 Cohere.

Compared with the 130 other AI sites in the library, Cohere sits with the 76% that stay light, and the group's most common heading face is Inter, where Cohere uses Cohere Text. The average AI homepage here has 7 sections; this one has 2.

Worth studying here: the hero, for how the headline, supporting line and primary action are stacked and the closing call to action, for contrast and copy length. Open any section for its palette, fonts and a build prompt you can hand to Claude, Cursor or v0.

More references: browse all websites.

More ai sites

Related searches

Sections captured: Hero, Call to action