Skip to content
FomoQuant

Fomo intelligence layer

Quantify the timeline.

Turn Fomo theses, users, token activity and market data into structured signals, scanners and backtests.

Open ScannerExplore Signals
Tokens
199
Theses · 24h
781
Active signals
2
Theses in. Scores out.
Live data
1Fomo activity
$QI
Meme
Theses 1h
3
Authors 1h
3

59 theses · 16 authors in 24h

2Features → drivers
  • Thesis velocity
    +520%
    +32 pts
  • Unique author growth
    +50%
    +6 pts
  • Acceleration
    87th pct
    +8 pts
  • Mention persistence
    21/24 h active
    +18 pts
3Score
Momentum
64/ 100

Confidence HIGHBased on 59 theses · 16 authors in 24h

momentum-v1.0
CONV
40
CROWD
36
EQ
—
NARR
—
DIV
+58
No active signal
Updated 48s agoOpen $QI

How it works

Every thesis leaves data behind.

FomoQuant continuously ingests authorized Fomo activity, joins it with market and onchain context, and turns it into point-in-time features, explainable scores and versioned signals. Nothing is calculated ad hoc in your browser.

  1. 01

    Sources

    • Fomo API — theses, authors, timing
    • Market — price, volume, liquidity
    • Onchain — holders, concentration

  2. 02

    Feature store

    • Point-in-time vectors per token
    • Velocity, breadth, concentration
    • Author quality as of T, never later

  3. 03

    Scores

    • Six scores + Smart Cohort
    • Drivers that sum to the score
    • Confidence and model version

  4. 04

    Signals

    • 10 versioned rules
    • Neutral reasons, thresholds, actuals
    • Historical analog statistics

  5. 05

    Outputs

    • Scanner and saved scans
    • Alerts and cohort convergence
    • API, stream, webhooks, backtests

  • Point in time

    Features at time T only use data available at T — no lookahead into prices, liquidity or later author history.

  • Explainable

    Every score ships with its drivers, raw values, confidence, sample size and model version.

  • Honest freshness

    When an upstream provider is delayed, pages say so — stale numbers are dimmed, never dressed up as live.

Scores

Six scores. Every number explains itself.

Each score is built from named drivers whose points add up to the value, with a confidence level, the sample it rests on and the model version that produced it. No black boxes.

Momentum

momentum-v1.0

Measures how fast Fomo attention on a token is accelerating right now.

  • Thesis velocity
  • Unique author growth
  • Acceleration
  • Mention persistence

0–100

Conviction

conviction-v1.0

Separates sustained attention from one-off noise.

  • Repeat authors
  • Active days
  • Author persistence
  • Revisit rate

0–100

Crowding

crowding-v1.0

Concentration of the last 24h of attention.

  • Top author share
  • Top 5 author share
  • Concentration
  • Thin breadth

0–100

Early Quality

early_quality-v1.0

Historical track record of the accounts posting early on a token.

  • Research score
  • Hit rate
  • Low false positives
  • Lead time

0–100

Narrative Velocity

narrative_velocity-v1.0

How fast the token's narrative (AI Agents, Gaming, DePIN, …) is spreading through Fomo.

  • New authors per hour
  • Narrative growth
  • Network spread
  • Cross-token mentions

0–100

Divergence

divergence-v1.0

Signed gap between social attention and market activity.

  • Social z
  • Price z
  • Volume z

−100 to +100

Plus Smart Cohort — whether a historically strong cohort is converging on the same token. Read every model definition →

Click a score. See exactly why it exists.

$QIsolana

Live data· computed 48s ago

Momentum
64/ 100
momentum-v1.0

Attention is accelerating.

DriverValuePoints
Thesis velocity
3 theses in the last hour vs a 7-day baseline of 0.40/h (velocity uses the 0.5/h floor)
+520%+32
Unique author growth
3 authors this hour vs 2 the hour before
+50%+6
Acceleration
3 vs 2 theses in the prior hour
87th pct+8
Mention persistence
21 / 24 hours active+18
Total 64
  • Thesis velocity percentile 98th
  • Unique author growth percentile 87th
  • Acceleration percentile 87th
  • Mention persistence percentile 100th
Confidence: HIGHConfidence: HIGH

Based on: 59 theses · 16 authors in 24h

Model momentum-v1.0 · sample size 59

Confidence is part of the number.

Early Quality · $QI

—/ 100
Confidence
Confidence: Insufficient data
Based on
0 qualifying authors · 0 historical calls
Model
early_quality-v1.0

Not enough resolved history among this token's early authors — shown as “—”, never estimated.

See every token's scores in the scanner

Signals

Signals describe. They don't prescribe.

10 versioned rules combine scores and features. Each one carries its reasons, the thresholds it passed and the historical analog statistics at fire time — statistical context, not a promise.

Active right now

Live data

$CYBERLEEK
False breakout risk
Momentum
51
Early Q.
—
Crowding
81
  • thesis velocity +399%
  • crowding high (81)
  • no early-author track record
36 historical analogs · median 24h -5.1% · worst -49.1%Details
$PUMPOWEEN
Crowding risk
Momentum
14
Early Q.
—
Crowding
85
  • crowding high (85)
  • top author wrote 67% of 24h theses
  • 1 account wrote half of 24 theses
3 historical analogs · median 24h -9.0% · worst -39.9%Details
Explore Signals

We say

  • Attention is accelerating.
  • Historically unusual author growth.
  • Social activity is leading price.
  • High concentration among 3 accounts.

We never publish

  • Trade calls or price targets
  • Promised outcomes
  • Multiples, hype and urgency
  • Unexplained black-box scores

Colours encode state and intensity only — accelerating, broad, quality, caution, cooling. Never a trade-direction colour.

  • Early accelerationThesis activity is accelerating early, with broad participation and historically strong early posters.
  • Broad attentionMany independent authors are posting; attention is not concentrated in a few accounts.
  • High-quality cohortSeveral historically consistent early posters are active on this token.
  • Crowding riskAttention is concentrated among a small number of accounts.
  • Social leading priceSocial activity is rising faster than price and volume.
  • Price leading socialPrice and volume are moving ahead of thesis activity.
  • Attention coolingThesis activity is decelerating versus the prior window.
  • Conviction buildingAuthors keep returning to this token across multiple windows.
  • Narrative breakoutThe token's narrative is spreading faster than its usual pace.
  • False breakout riskA spike in activity with weak breadth and low historical quality — similar setups often faded.

Scanner & rule builder

Scan the whole network with one rule.

Filter every token on any score or feature. Start from a preset or build a rule without code, then save it as a scan or turn it into an alert.

WhenConviction>70andRepeat author ratio>30%
ThenShow in scannerAlert me
  • Early Attention
  • Broad Momentum
  • High Conviction
  • Low Crowding
  • Smart Cohort
  • Cooling Fast
  • Narrative Breakout

8 of 198 tokens match High Conviction

Live data

TokenMomCrowdAuthors 1hSignal
$PUMPOWEEN
Other
11840Crowding risk
$招财猫
Other
4700—
$AGENCY
AI Agents
6780—
$MEMESTOCK
Other
——0—

Snapshot 18m ago · Free plan scanner is delayed 15 min; Pro and Team are realtime.

Open ScannerRule fields reference

Backtests & event studies

Backtests that show the downside.

Never only positive metrics. Every result carries its sample size, confidence and date range, the full return distribution and the drawdowns along the way — computed strictly point in time.

Event study

What happens after thesis velocity triples?

Event: Thesis velocity > 3× baseline

Confidence: MEDIUM30 events with 24h outcomes · 27 tokens · 2026-10-08 → 2026-10-10
Samples
30
Median 1h
-0.1%
Median 24h
+2.0%
Median max DD
-17.9%
within 24h
Positive 24h
53%

Return 24h after the event

N = 30 · Positive after 24h 53% (95% CI 36–70%)

Median
+2.0%
CI -8.7% … +8.0%
Trimmed mean
+0.6%
10% each side
Mean
+14.9%
σ 99.8%
Range
-76.6% … +523%
Return 24h after the event histogram30 samples in 9 bins from -76.6% to +523%. Tail bins: 1 above +63.5%.median-76.6%+523%0%+63.5%-50%+50%
Return 24h after the event box plotMedian +2.0%, interquartile range -12.7% to +10.7%, whiskers -40.3% to +45.5%, 3 outliers (min -76.6%, max +523%).-40.3%+45.5%+2.0%

2 below -59.2% (min -76.6%) · 1 above +64.4% (max +523%)

p5
-52.2%
p10
-37.1%
p25
-12.7%
p50
+2.0%
p75
+10.7%
p90
+40.9%
p95
+43.8%
  • Short history (2.1 days of snapshots)
  • 19 events too recent to resolve (horizon ends after now) — excluded from statistics

Live data· computed 46m ago

Full study

Max drawdown within 24h

N = 30 · median -17.9% · worst -78.2%

Max drawdown distribution30 samples in 9 bins from -78.2% to -0.4%.median-78.2%-0.4%-60%-40%-20%

No leakage, no survivorship

  • Features at event time T use only data available at T
  • Entry is the first price at or after the event — never before
  • Dead and failed tokens stay in the sample
  • Non-overlapping events per token (cooldown = horizon)
  • Medians, trimmed means, Wilson and bootstrap intervals
  • Warnings for small, concentrated or short samples
Methodology
Run a backtestEvent studies

FomoQuant provides analytics and historical/statistical context, not financial advice or guaranteed predictions.

Cohorts & narratives

Who is converging, and which stories are spreading.

Track groups of Fomo users — your own or algorithmic cohorts with minimum sample sizes — and watch attention rotate between narratives hour by hour.

24h rotation · attention share

Live data· computed 53m ago

  • Other77% -3.1 pts
  • Meme7% +1.1 pts
  • AI Agents6% -1.0 pts
  • Consumer4% +1.7 pts
  • Gaming2% +1.1 pts
  • Launchpads2% -0.3 pts
  • DeFi1% +0.8 pts
  • Infra & L1/L21% +0.1 pts
  • RWA0% -0.1 pts
  • DePIN0% -0.1 pts

Share of all theses in the last 24h; the change is in percentage points versus the previous 24h. Colours identify narratives, not direction.

Narrative map & heatmap

Cohort convergence

No convergence in the latest window. Convergence needs at least max(3, min(5, ⌈15% of members⌉)) distinct members posting on the same token within 60 minutes.

Algorithmic cohorts

  • Historically Early — Authors with a research score ≥ 70 whose calls historically preceded major moves by a median of 12 hours or less (≥ 15 resolved calls).
  • AI Specialists — Authors whose resolved AI calls score ≥ 70 on sector strength (at least 8 resolved AI calls).
  • Gaming Specialists — Authors whose resolved Gaming calls score ≥ 70 on sector strength (at least 8 resolved Gaming calls).
  • High-Consistency Posters — Authors whose weekly hit rates vary little (consistency ≥ 75) across at least 20 resolved calls.
  • High False-Positive Accounts — Authors where ≥ 60% of resolved calls saw no major move and a negative 3-day return (≥ 15 resolved calls). Tracked as a contrast cohort.
Explore cohorts

Developer API & SDK

Raw Fomo activity in. Structured signals out.

The same numbers the app shows, over a REST API, a WebSocket stream, signed webhooks and a typed, zero-dependency TypeScript SDK.

  • GET/v1/quant/token/{token}Scores, drivers, confidence, signals
  • GET/v1/quant/signalsSignal feed with filters
  • POST/v1/quant/scanRun scanner rules
  • GET/v1/quant/users/{handle}Point-in-time author metrics
  • GET/v1/quant/cohorts/{id}Members, activity, convergence
  • POST/v1/quant/backtestsQueue a point-in-time backtest
  • WS/v1/streamsignal.created, score.updated, …
  • snake_case JSON with model_versions and freshness
  • Scoped, hashed keys · per-key rate limits
  • Webhooks signed with HMAC-SHA256
  • Typed errors, automatic 429 retry
QuickstartAPI reference 
import { FomoQuant } from "@fomoquant/sdk";

const fq = new FomoQuant({ apiKey: process.env.FOMOQUANT_API_KEY, baseUrl: "https://fomoquant.vercel.app" });

const token = await fq.tokens.get("$QI");
const m = token.score_details.momentum;
console.log(m.value, m.confidence.level, m.model_version);
for (const d of m.drivers) console.log(d.label, d.display, d.points);

// every active signal, across pages
for await (const s of fq.signals.iterate({ status: "active" })) {
  console.log(s.token.symbol, s.label, s.why);
}

The Response tab is the real current snapshot of $QI, serialised exactly as the API returns it; long sections are collapsed.

Principles

Measure, don't predict.

FomoQuant tells you what the Fomo network is actually doing — never what to trade. The moat is authorized data, historical outcomes, reproducible scores and honest statistics.

  • Every number is explainable

    Score → drivers → raw values → confidence → model version. If we can't show why a number exists, we don't show it.

  • Confidence everywhere

    Sample size, date range and a confidence level travel with every score, backtest and event study. Ten samples are labelled as ten samples.

  • Point in time, always

    History uses only what was knowable then. Editing a profile today can't rewrite yesterday's scores; dead tokens stay in the record.

  • Live mainnet data

    Every number comes from real Fomo theses on mainnet chains — Robinhood Chain, Solana, Base, BNB Chain, Ethereum and Arc — and real market data. Nothing is simulated.

  • Never fabricated

    When a provider lacks a field, it is “—” with the reason. When inputs are delayed, the page says so.

  • Authorized sources only

    Fomo data arrives through the authorized API — no scraping, no private endpoints. FomoQuant is independent of FOMO Labs.

Our promise

Things FomoQuant will never build

  • Auto trading
  • Wallet trading
  • Copy trading
  • Leverage
  • Prediction guarantees
  • Portfolio custody
  • AI chat trader
  • Token launchpad
  • Social feed clone

Pricing

Start free. Upgrade for realtime.

Browsing needs no account and no wallet. Billing is not live yet — Pro and Team are enabled on request.

Compare plans

Free

Explore live signals and scores, a delayed scanner and the live API at 60 req/min.

$0

No card, no wallet needed

  • 15-min delayed scanner · 7 days of history
  • 3 alert rules · In-app, Email
  • 2 saved scans
  • Event studies
  • API: live keys · 60 req/min
Start free

Pro

Most capable

Realtime scanner, custom rules, your own cohorts, backtests and a higher API rate limit.

$49/ mo

Planned price · billing is not live yet

  • Realtime scanner · 1 year of history
  • 50 alert rules · In-app, Email, Telegram, Discord
  • 50 saved scans
  • Custom rule builder
  • Your own cohorts
  • Backtests
  • Event studies
  • API: live keys · 600 req/min
Request access

Team

Everything in Pro, plus a shared workspace for scans, cohorts and API usage.

$199/ mo

Planned price · billing is not live yet

  • Realtime scanner · 1 year of history
  • 200 alert rules · In-app, Email, Telegram, Discord
  • 200 saved scans
  • Custom rule builder
  • Your own cohorts
  • Backtests
  • Event studies
  • API: live keys · 2,000 req/min
  • Shared team workspace
Request access

Stop reading the feed.
Measure it.

Theses in. Signals out. Every number with its reasons, its sample size and its model version.

Open ScannerRead the docs