# Answer compare

A React component in [Beamline](https://beamline.io/)'s [AI & agents](https://beamline.io/components/ai) category. Live demo: https://beamline.io/components/answer-compare

Two answers to the same prompt side by side, and a person's verdict: A is better, a tie, B is better, or both are bad, with the reasons that decided it. Answers stream in their own columns with time, tokens and cost under each; in a blind test the models stay hidden behind A and B until the verdict, then turn over to show who wrote what. Narrow rooms switch between the two instead of squeezing them.

## Use it for

- Choosing between two models, prompts or versions on the same input (a playground, an eval queue, a prompt review).
- Collecting a person's preference with reasons, blind or named.

## Not for

- One answer in a conversation ([message](https://beamline.io/components/message), [response](https://beamline.io/components/response)).
- Comparing two pictures ([image-compare](https://beamline.io/components/image-compare)).
- Comparing plans or prices ([plan-comparison](https://beamline.io/components/plan-comparison)).

## Anatomy

- prompt: the input both answered, one line that opens to the whole
- columns: A and B, each with its name (A, B, or the model), the answer (streamed markdown, or your own drawing) and its facts: time, tokens, cost (not reported is said)
- narrow room: a switch between A and B above one column, each answer keeping its scroll
- verdict bar: A is better, Tie, B is better, Both are bad; the chosen one pressed, the better column ringed and the other quieter
- reasons: after a verdict, chips for why (More accurate, Clearer, Shorter, Follows the instructions, Safer…) and an optional note; Submit sends
- blind: model names hidden until the verdict; they turn over to show

## Variants

- **blind**: true, false

## States

- streaming
- ready
- chosen
- reasons
- submitted
- error (an answer that failed says so in its column)

## Keyboard

- The verdict buttons are a group: ← → move, Enter or Space chooses; 1, 2, 3 and 4 choose from anywhere in the element
- The columns scroll with the page or on their own; with sync on they scroll together
- In a narrow room the switch is a segmented control

## Motion

Each answer streams in its column at the response's even pace. Choosing rings the better column and quiets the other (a cross-fade), the reasons grow under the bar, and in a blind test the names turn over one after the other. Submitting folds the reasons into a line that says what was sent. Reduced motion: the names change in place, nothing turns.

## Props

| Prop | Type | Default | Description |
| --- | --- | --- | --- |
| `prompt` | `ReactNode` | — |  |
| `answers` (required) | `[CompareAnswer, CompareAnswer]` | — |  |
| `value / defaultValue / onValueChange` | `CompareVerdict \| null` | — |  |
| `onSubmit` | `(verdict: CompareVerdict) => void \| Promise<void>` | — | With it, reasons and Submit follow the choice; a promise shows progress and a refusal keeps the choice. |
| `reasons` | `string[]` | — | The reason chips. Default: More accurate, Clearer, More complete, Shorter, Follows the instructions, Safer. |
| `blind` | `boolean` | — | Hide the models until the verdict. |
| `syncScroll` | `boolean` | — | The columns scroll together. Default true. |
| `currency` | `string` | — | Default "$". |
| `className / classNames / style / ref` | `root` | — |  |

## Built from

- response
- button
- [chip-group](https://beamline.io/components/chip-group)
- [segmented-control](https://beamline.io/components/segmented-control)
- textarea

## Dependencies

`lucide-react`

## Import

```tsx
import { AnswerCompare } from "@/components/answer-compare/answer-compare";
```

## Get it

Answer compare is part of Beamline: 200+ components and 50+ complete screens for React that your coding agent (Claude Code, Codex, Cursor or any MCP client) installs into your app through Beamline's MCP server, as a ready-built package or as plain React source you can change. $49 launch price, one payment, no subscription, lifetime updates, one licence for your whole team. [Get Beamline](https://beamline.io/checkout?pack=complete) · [Connect your agent](https://beamline.io/connect)

## More in AI & agents

- [Image generation](https://beamline.io/components/image-generation): The frame an AI image is generated into: a field of brand-hued pixels breathes with the progress while the model works, then the image arrives —…
- [Flowchart](https://beamline.io/components/flowchart): A workflow as steps on a dotted canvas: a trigger, conditions that branch into paths side by side, actions and waits, read top to bottom; during a run…
- [Run trace](https://beamline.io/components/run-trace): One agent run as a waterfall: every step (model calls, tools, retrieval, checks, sub-agents) nested under the step that started it, each a bar on a…
- [Response](https://beamline.io/components/response): An assistant's answer as streamed markdown in the system's reading type: bursty tokens flow in at an even pace, each new word fades in through a short…
- [Chat thread](https://beamline.io/components/chat-thread): People talking to people: messages grouped by person and day, reactions, delivery and read receipts that glide to the newest message read, who is typing…
- [Model picker](https://beamline.io/components/model-picker): Choose the AI model for a request by what it can do and what it costs: a trigger with the model's name opens a searchable list grouped by provider, with…
- [Task rows](https://beamline.io/components/task-rows): Background work an agent is doing, one row per task: status as a glyph and a word, what it produced, how long it has run or took, a progress bar while…
- [Conversation](https://beamline.io/components/conversation): The scrolling column of a conversation: follows new content while you are at the bottom, stops when you scroll up to read, and offers a way back; shared…
- [Context cards](https://beamline.io/components/context-cards): The knowledge an agent retrieved before it answered: one card per chunk with what it is about, the text it read with the search words marked, how…
- [Prompt suggestions](https://beamline.io/components/prompt-suggestions): Ready-made prompts a person can start from or follow up with: chips, cards or a list; a picked one lifts off and flies into the composer as its text…
- [Plan](https://beamline.io/components/plan): An agent's plan: numbered steps with live status, progress, and Approve / Edit / Reject when it needs sign-off; approving shows progress, and a refusal…
- [Approval card](https://beamline.io/components/approval-card): The agent stops and asks before it acts: what it wants to do, exactly what that changes, any choices it needs, and Approve or Reject; both run your…
- [Tool call](https://beamline.io/components/tool-call): One tool call in an agent's run: status, what it did and a short result on one quiet row that opens to the exact input and output; stops for permission…
- [Recommendation card](https://beamline.io/components/recommendation-card): One suggestion from the agent that the person may take or leave: what it suggests, why, the expected effect and how sure it is (a meter and a word…
- [Citations](https://beamline.io/components/citations): Inline numbered citations that preview their source on hover or focus, plus a numbered source list; streamed answers flow in at an even pace with the…
- [Reasoning](https://beamline.io/components/reasoning): The agent's trace, fed by your stream: a shimmering label and the seconds so far while it works, then "Thought for N seconds"; steps that check off…
- [Attachment](https://beamline.io/components/attachment): A file tile: preview or typed glyph with the type named, name, size, upload progress, error with retry, and a keyboard-reachable remove.
- [Prompt bar](https://beamline.io/components/prompt-bar): The agent composer: a field that grows, @ to add a source, / for a command, a model picker, files and dictation, and one button that sends with progress…
- [Selection actions](https://beamline.io/components/selection-actions): Select a passage and hand it to the agent: a bar over the selection with quick actions, your own instruction and More; while the agent works it says so…
- [Message](https://beamline.io/components/message): One turn in a conversation with an agent: the person's words in a quiet bubble, the agent's answer on the page with its orb mark, trace, tool calls and…