The receipts

Live receipts.

We detect pump.fun graduations before the bonding curve completes. Every prediction logged with a hash before the outcome — receipts anyone can verify. Three numbers, read them together:

Median runway · ≥0.70 calls

—s

between our call and the bonding curve completing

Median over the last 90 days · sample: — resolved high-confidence graduations · bot-actionable in real time

Calibrated accuracy · ≥0.70

…

of those calls actually graduate within 24h

Forward — …; every prediction hashed before outcome

Sustain · 30m post-bond

…

of graduated mints held ≥80% of grad price 30 min later

n=… resolved post-bond outcomes

The three numbers measure different things and shouldn't be averaged. Median runway is the time between our ≥0.70 confidence alert and the bonding curve actually completing — the window a fast trader or sniper bot has to execute a buy on the curve before migration. Calibrated accuracy is the share of those calls that actually graduate within 24h — when the model says ≥0.70, that's the reality. Sustain 30m post-bond asks "of mints that did graduate, how many held value 30 min later?" — sourced independently from on-chain DEX prices, the question a trader actually has, since graduation alone is not a profit thesis. Pre-launch audit on /verdict — we caught and corrected our own measurement bug before launch; that's the receipts chain working.

By confidence band

The model is honest about uncertainty. Lower-confidence calls graduate at lower rates — exactly as predicted. The Telegram bot only fires at ≥70%.

If we say	Actual graduation rate	Sample size
loading…

SYSTEM STATUS

CALIBRATED · STABLE

verify any prediction →

runner_prob fields — directional, recalibration pending

The runner_prob_2x/5x/10x_from_now fields exposed at /api/v1/probe and /api/live are currently directional: mints with higher runner_prob do hit higher rates, but the absolute probability is over-stated by roughly 11-13 percentage points on high-confidence bins (≥0.5). The saturation case (kNN reports 1.0 because all neighbors hit) is the loudest miss — runner_prob_5x_from_launch at predicted 0.99 has actual rate around 0.29.

Magnitude recalibration is in progress via the existing apply_calibration infrastructure (the same self-correcting curve grad_prob uses). Until recalibration is verified, treat runner_prob fields as a ranking signal, not as a literal probability. Consumers making sizing decisions on the absolute number should discount by ~12pp at the high end. The full audit (/api/scope documents the field, n=89,077 sample) is intentionally surfaced here rather than hidden — same discipline as the warming label on the live rate above.

raw JSON: /api/accuracy · NFA · DYOR · prediction model output, not financial advice