​
What is the Neynar user score?

Neynar user score is generated based on user behavior on the platform. It scores between 0 and 1 and reflects the confidence in the user being a high-quality user. Users can improve their scores by having high-quality interactions with other good users. Scores update weekly. If you want to see your score as a user, you can use the Neynar Explorer and insert your fid in the search bar. Sample url explorer.neynar.com/<fid>.

​
Scores are also available onchain, see Address <> user score contract

​
Interpreting the score

The score is not proof of humanity. It’s a measure of the account quality / value added to the network by that account. It’s capable of discriminating between high and low quality AI activity. E.g. agents like bracky / clanker are bots that have high scores while accounts posting LLM slop have lower scores. You can see a distribution of users across score ranges on this dashboard. A screenshot from Dec 5, 2024 is below.
Neynar user score distribution
We recommend starting with a threshold around 0.5 and then changing up or down as needed. As of Dec 5, 2024, there are:
  • ~2.5k accounts with 0.9+ scores
  • ~27.5k accounts with 0.7+ scores
Hence, starting with a very high threshold will restrict the product to a tiny user cohort. Developers should assess their own thresholds for their applications (Neynar does not determine thresholds in other apps). Scores update at least once a week, so new users might take a few days to show an updated score. If the user has not been active for a while, their scores will be reduced.

​
Fetching the score for development

​
Getting the score on webhook events

If you’re using Neynar webhooks to get data on your backend, you might want to separate high-quality data from low-quality data. A simple way to do this is to look at the neynar_user_score inside each user object. 
user: {
	fid: 263530,
	object: "user",
	pfp_url: "https://imagedelivery.net/BXluQx4ige9GuW0Ia56BHw/68c1cd39-bcd2-4f5e-e520-717cda264d00/original",
	profile: {
		bio: {
			text: "Web3 builder"
		}
	},
	username: "m00n620",
	power_badge: false,
	display_name: "tonywen.base.eth",
	experimental: {
		neynar_user_score: 0.9 // THIS IS THE SCORE
	},
	verifications: [
		"0xc34da1886584aa1807966c0e018709fafffed143"
	],
	follower_count: 29,
	custody_address: "0x22c1898bddb8e829a73ca6b53e2f61e7f02a6e6d",
	following_count: 101,
	verified_accounts: null,
	verified_addresses: {
		eth_addresses: [
			"0xc34da1886584aa1807966c0e018709fafffed143"
		],
		sol_addresses: []
	}
}

​
Fetching the score on API calls

If you’re using APIs, you will see the same score in all user objects across all Neynar API endpoints. Try the following endpoints on our docs pages and look at the user object to see this in action:
  • User by FIDs to see this when fetching user data by fid
  • By Eth or Sol addresses If looking to restrict activity to a specific cohort of users, you can check user score on any Neynar API endpoint and then allow them to take actions as appropriate.

​
Pulling the scores from hosted SQL

Neynar SQL playground has user scores available and you can pull from there for larger analysis as needed. Reach out if you don’t yet have access to it.
Neynar user score in SQL

​
Pulling score data onchain

See here for details. Reach out to us on Slack if you need the data to be updated.

​
Report errors

If you know a score misrepresents a user, that’s helpful information we can use to label our data. Please send feedback to @rish on Warpcast DC or Slack .

​
FAQs

​
1. How often do the scores update?

There’s two different things to note here:
  • (a) The algorithm runs weekly and calculates scores for accounts on the network based on their activity
  • (b) The algorithm itself is upgraded from time to time as activity on the network shifts. You can read about one such update in Retraining Neynar User Score Algorithm

​
2. As a user, how can I improve my score?

The score is a reflection of account activity on the network. Since we have introduced this score, a few users have written helpful guides on how to contribute to the network in a positive manner:
  • see from @ted here
  • longer write up from @katekornish (now turned into a mini app by @jpfraneto) here