Uniswap v1: Automated Market Maker on Ethereum

Author: Hayden Adams
Year: 2018 (cited as 2018 by Uniswap v2 Core)
Project: Uniswap v1
Source status: Author-hosted draft; the HackMD page is visibly marked “WORK IN PROGRESS”
Document license: Not stated on the linked HackMD document

Original paper Original paper
Directories Ethereum Whitepapers & Research

This page summarizes the historical design of Uniswap v1; it is not a guide to using legacy contracts. It is for education, not financial advice. Consult the original sources before making decisions.

Uniswap v1 was an early automated market maker (AMM) design on Ethereum. Hayden Adams’s Uniswap Whitepaper draft describes exchange contracts for ETH and individual ERC-20 tokens, with prices changing as each pool’s reserves change. The source labels itself as work in progress, so this page treats it as a historical design draft rather than a final specification.

The factory and ETH/ERC-20 pairs

The Uniswap factory registers an exchange contract for an ERC-20 token. Each exchange holds ETH and one ERC-20 asset, and traders swap against those reserves. Because ETH served as a common intermediate asset, a trade between two ERC-20 tokens could route through two pools: from the first token to ETH, then from ETH to the second token. The archived official v1 documentation describes this factory and routing model.

Pricing and fees

Each pool uses a constant-product pricing rule, commonly written as x × y = k. When a trader adds one asset and withdraws the other, the reserves change and the quoted price moves. The effect depends on the trade size relative to the pool: larger trades relative to available reserves generally create more slippage.

The v1 draft describes a 0.3% trading fee that remains in the pool’s reserves and contributes to the value liquidity providers can withdraw. Traders can set minimum output or maximum input bounds and a transaction deadline. These are historical design details; they should not be assumed to describe every current Uniswap version.

Liquidity and pool tokens

A liquidity provider deposits ETH and an ERC-20 token into an exchange. The contract issues pool tokens to track the provider’s share of the reserves. When liquidity is withdrawn, those pool tokens are burned and the provider receives a proportional share of the available reserves. The asset mix can change as trades occur, so fees do not remove price risk.

V1 limits and what changed in v2

Requiring ETH in every public pair limited direct pair choices and could make ERC-20-to-ERC-20 trades pass through two pools. A pool’s spot price also should not automatically be treated as a safe oracle for another contract: trading can move the price, so manipulation risk matters.

The official Uniswap v2 Core paper (2020) describes later changes, including direct ERC-20/ERC-20 pairs, cumulative price tracking for estimating a time-weighted average price, and flash swaps. Those are v2 features; they should not be attributed to v1.

Source status and limits

The linked HackMD source is titled “Uniswap Whitepaper” and visibly says “WORK IN PROGRESS.” The v2 Core paper cites Hayden Adams’s work as “Uniswap v1 Core” (2018), but the draft status on the original page still matters. The GPL-3.0 license shown for the archived v1 documentation repository does not establish a separate license for the HackMD paper, so no license is assigned here.

Related Terms

See Also

Sources