What Is Hash Rate?
Hash rate measures the total computational power being dedicated to mining a Proof-of-Work blockchain. It represents the number of hash computations the network (or an individual miner) can perform per second. Higher hash rate means more security — an attacker would need to control a majority of the network’s total hash power to execute a 51% attack, so more hash rate means a more expensive and difficult attack.
Hash rate is the fundamental metric for understanding the health and security of any PoW blockchain. When you hear that “Bitcoin’s hash rate reached an all-time high of 700 EH/s,” it means the combined mining hardware on the Bitcoin network is performing 700 quintillion hash calculations every second. That number is almost incomprehensibly large, and it is exactly what makes Bitcoin the most secure computer network ever created.
Units of Measurement
Hash rate is expressed using standard SI prefixes, though the scale is staggering compared to most other fields:
| Unit | Abbreviation | Hashes Per Second |
|---|---|---|
| H/s | Hash per second | 1 |
| KH/s | Kilohash | 1,000 |
| MH/s | Megahash | 1,000,000 |
| GH/s | Gigahash | 10^9 |
| TH/s | Terahash | 10^12 |
| PH/s | Petahash | 10^15 |
| EH/s | Exahash | 10^18 |
| ZH/s | Zettahash | 10^21 |
Bitcoin network hash rate is measured in Exahashes per second (EH/s). As of mid-2025, Bitcoin’s hash rate has exceeded 700 EH/s. This means the network performs more hash computations in a single second than all the world’s general-purpose computers could perform in years.
Ethereum (before the Merge) was measured in Terahashes per second (TH/s). At its peak, Ethereum’s PoW hash rate reached approximately 1,100 TH/s before transitioning to Proof-of-Stake in September 2022.
Litecoin hash rate is typically measured in TH/s, while smaller networks like Dogecoin and Monero use GH/s or MH/s.
How Hash Rate Is Estimated
You might wonder: how does anyone know the total hash rate of a decentralized network? No single entity tracks every miner. Instead, hash rate is inferred from the difficulty and block production rate:
- The protocol adjusts difficulty to target a specific block time.
- If blocks are being found at exactly the target rate, we can estimate:
Estimated Hash Rate = Difficulty × 2^32 / Target Block Time - If blocks are being found faster than the target, actual hash rate is higher than the estimate.
This means hash rate estimates have a lag — they reflect the hash rate from the previous difficulty period. Additionally, block times are random (Poisson distribution), so there is natural variance in the estimates. Hash rate charts typically use a moving average (7-day or 30-day) to smooth out this noise.
Mining Hardware and Hash Rate Efficiency
Different types of mining hardware produce vastly different hash rates:
ASICs (Application-Specific Integrated Circuits): Hardware designed exclusively for mining a specific algorithm (SHA-256 for Bitcoin). Modern Bitcoin ASICs (e.g., Bitmain Antminer S21, MicroBT WhatsMiner M60) produce 150-250 TH/s per unit. ASICs dominate Bitcoin mining because they are orders of magnitude more efficient than GPUs or CPUs for SHA-256 hashing.
GPUs (Graphics Processing Units): General-purpose parallel processors originally designed for gaming and later adopted for cryptocurrency mining. GPUs are flexible — they can mine different algorithms (Ethash, RandomX, Autolykos) and are useful for AI workloads. A high-end GPU (NVIDIA RTX 4090) produces roughly 100-200 MH/s on memory-hard algorithms like Ethash. GPUs were the backbone of Ethereum mining before the Merge.
CPUs (Central Processing Units): General-purpose processors. Their hash rate for most mining algorithms is extremely low (a few KH/s to MH/s). However, Monero’s RandomX algorithm is specifically designed to be CPU-friendly and ASIC-resistant, making CPUs the primary mining hardware for Monero.
FPGAs (Field-Programmable Gate Arrays): Reprogrammable hardware that sits between ASICs and GPUs in terms of efficiency and flexibility. Some mining operations use FPGAs for specific algorithms where they offer better energy efficiency than GPUs.
Hash Rate and Network Security
Hash rate is directly tied to network security through the cost of a 51% attack:
- Attack cost = Hardware cost + electricity cost to sustain >50% hash rate for a meaningful period.
- For Bitcoin at 700 EH/s, an attacker would need to acquire and operate >350 EH/s of SHA-256 ASICs — equivalent to billions of dollars in hardware and tens of millions in daily electricity.
- For smaller PoW networks with hash rates in the MH/s or GH/s range, a 51% attack is much cheaper. This is why networks like Ethereum Classic and Bitcoin SV have suffered repeated 51% attacks.
The security economics are clear: hash rate is expensive to acquire and maintain, which is what makes PoW blockchains resistant to attacks. The cost of attacking Bitcoin is so high that no rational actor would attempt it unless the potential gain (e.g., a double-spend of billions of dollars) justified the enormous hardware and energy expenditure.
Mining Profitability and Hash Rate
Mining profitability depends on several factors related to hash rate:
- Revenue per hash: Block reward + transaction fees, divided by total network hash rate. As more miners join, each miner’s share of rewards decreases.
- Energy cost per hash: Measured in joules per terahash (J/TH). More efficient ASICs have lower J/TH ratings, meaning they produce more hashes per unit of electricity.
- Profit margin: Revenue per hash minus energy cost per hash. When Bitcoin’s price drops or difficulty rises too fast, miners with older, less efficient hardware become unprofitable and must shut down.
The concept of “hashprice” (market value of hash rate) emerged as a way to standardize mining economics. Hashprice is calculated as the daily revenue per terahash per second, typically measured in dollars. It fluctuates with Bitcoin’s price, network difficulty, and transaction fee levels.
Common Pitfalls
- Confusing individual hash rate with network hash rate: Your ASIC produces 200 TH/s; the network is at 700 EH/s. You are a tiny fraction of the total.
- Assuming higher hash rate means higher mining revenue: Individual mining revenue depends on your share of total hash rate, not your absolute hash rate. If you double your hash rate but the network hash rate also doubles, your revenue stays the same.
- Ignoring energy efficiency: A miner with 300 TH/s at 20 J/TH is more profitable than one with 200 TH/s at 15 J/TH if the hash price justifies the electricity cost.
- Reading too much into short-term hash rate changes: Because hash rate is estimated, daily fluctuations are noisy. Focus on 30-day or 90-day trends for meaningful analysis.
- Comparing hash rates across algorithms: Bitcoin’s 700 EH/s (SHA-256) cannot be directly compared to Monero’s 3 GH/s (RandomX). Different algorithms have completely different computational costs per hash.