SN21 — the daily stream

The calendar.

Each result settles on its change date, plus one day, plus the horizon, plus a two-day settling window: 10, 17 and 31 days after the change. The table is the launch calendar published on 5 August 2026, when the first live basket covered Monday 3 August. From 18 August a fresh 7-day cohort settles every day, each basket on its own clock.

Burn figures are a published plan and may be adjusted, up or down, to protect alpha value for holders. The alpha hold is a minimum balance, not a purchase: there is no upfront buy-in, and earnings can fill it.

How a miner joins.

1. 1

   Register

   Register your hotkey on the subnet and bind an ed25519 signing key. One time.

2. 2

   Train

   Train on the published episodes — the inputs — and their settled outcomes — the labels — or the pre-joined training bundle. Outputs are P10/P50/P90 distributions at 7, 14 and 28 days: calibrated ranges, not point estimates.

3. 3

   Containerise

   Package the model as a container that reads episodes on stdin and writes predictions on stdout. No network, 1 GB of memory, 15 CPU-minutes per basket.

4. 4

   Commit the digest

   Push the image and commit its digest on chain. The subnet pulls that exact digest, admits it through the backtest gate, and runs it against every daily basket.

You do not post predictions. Your container is executed by the operator against each day's basket, and its output is locked before any outcome exists.

The step-by-step guide, with every command, is the miner quickstart; the training guide walks the model side end to end — join the data, train, containerise, and score yourself before you ship. Both are in the public repository.

Where the data is.

Everything you train on is public and lives in the repository. Episodes are the inputs you predict on; outcomes are the measured labels, published after each horizon settles. They pair one-to-one by epoch id. Start with whichever is easier: the raw corpus, or the pre-joined bundle.

• Published corpus

  Inputs and labels as separate files, paired by epoch id.

  data/episodes · data/outcomes

• Training bundle

  Settled episodes, pre-joined, all three horizons — one JSON object per line.

  Download .jsonl · release notes

• Training guide

  The full path: join, train, export weights, containerise, self-score.

  read on this site · in the repo

The bundle is held out from scoring, so training on it cannot leak a score. As the daily stream settles, each scored day is also published in full as a signed receipt you can train against — see verifying a day. The training guide names the exact commit each copy was taken from.

How big each day's basket is.

Every episode in a basket is one real change a model has to predict. This is the number of episodes in each day's basket since the first live one, updated daily.

Does it work?

The honest test of a prediction subnet is whether the best model gets better, and whether it holds up as the horizon lengthens. This is that record, week by week, and it is drawn from the same signed receipts a miner uses to check their own scores.

You can check the arithmetic.

Every scored day is published as a signed receipt — the settled outcomes used, every prediction exactly as the model produced it, each score's components — and the feed is anchored on chain. You can recompute your own scores from it and check they match ours. How to verify a day →

Canonical documents.

Each is mirrored on this site for reading and linking, and each mirror names the commit it was taken from. The public repository is the source of truth; where the two disagree, the repository and the chain win.

• Why daily

  docs/SN21_WHY_DAILY.md

• Transition plan

  docs/SN21_TRANSITION_PLAN.md

• Scoring

  docs/SN21_SCORING.md

• Rewards

  docs/SN21_REWARDS.md

• Staking

  docs/SN21_STAKING.md

• Model specification

  docs/MINER_MODEL_SPEC.md

• Training

  docs/SN21_TRAINING.md

• Verifying your score

  docs/SN21_VERIFYING.md