Decentralized GRPO on Bittensor SN81: verified frontier rollouts, validator-authoritative training, and public evidence.
What miners actually race on — the mission definition, its tasks and scoring, and the mission-score board. Not the same thing as emission share.
Mission home: reliquadotai/reliquary
Mission source failing since 4d ago — showing last good from 4d ago.
Round timeline: submissions — → evaluation #49,566 → scoreboard deadline #49,566 → round end —.
| Rank | Miner | Mission score |
|---|---|---|
| #1 | 5CkdyTX…KsPPF | 0.01 |
| #2 | uid #77 | 0.01 |
| #3 | 5CCnVDz…cqbEY | 0.01 |
| #4 | uid #139 | 0.00 |
| #5 | 5EqU239…MuDSm | 0.00 |
Read 4d ago from www.reliqua.ai/api/miners. Mission scores are the subnet's own scoring; emission share is the on-chain payout — different numbers, different meanings.
What the validator network actually scored this epoch — read from the chain's Incentive vector, normalized on chain.
These are each miner's normalized incentive from the chain's Incentive vector — the only race score Bittensor stores. It is a share of the validators' consensus weight: all rows sum to 1, so uid #56's 0.802 on Minos means it holds 80.2% of the validated weight — and since emission follows incentive exactly, it also takes ~80% of the miner-side payout. There is no absolute score (an “87/100”) on chain; numbers like that live in each subnet's own off-chain platform. Uids absent from the list received zero validated weight this epoch.
Absolute submission scores (an “87/100”) are computed by validators off chain and never published on chain — subtensor no longer even stores a pre-normalization Scores vector, so the incentive shares above are the only race score that exists on chain. When a subnet publishes scores through its own mission API, they appear in the mission panel above instead.
live chain reads
scores read 4d ago · block 9,485,344 · epoch 26,348
and what they can't
Incentive is the score the validator network converged on for each uid this epoch — it's what pays: emission follows it. It is normalized on chain to sum to 1, so it reads as a share of validated weight. Hidden per-validator test sets (e.g. a private eval split) are computed off chain and never published — the chain no longer even stores a pre-normalization Scores vector — so taonets shows exactly what the chain shows, no reconstruction.
did the race change? the head commit is the probe
The watch hasn't confirmed this repo in 4d ago — the head below may be behind GitHub. Treat the race as possibly moved until the watch is live again.
fix(orders): the failure reconcile reads verdicts in parallel, behind the audit (#497)
The head commit is the atomic “did the race move” probe: any push to the default branch — new tasks, benchmark changes, scoring edits — lands here on the next watch pass. The confirmed-ago line above proves the watch is actually running.
archived per epoch since taonets started tracking (epoch 25,629)
| Epoch | Block | Time | Winner uid | Winner coldkey | Race score | Top incentive | Qualifier mean incentive | Qualifiers |
|---|---|---|---|---|---|---|---|---|
| Nov 11 | ||||||||
| 26,348 | 9,485,344 | Nov 11, 20:11 | #237 | 5GnpQR4…mkAQT | — | 0.915 | 0.009 | 116 |
| 26,347 | 9,484,894 | Nov 11, 18:41 | #237 | 5GnpQR4…mkAQT | — | 0.862 | 0.008 | 120 |
| 26,346 | 9,484,592 | Nov 11, 17:41 | #237 | 5GnpQR4…mkAQT | — | 0.862 | 0.008 | 120 |
| 26,345 | 9,484,173 | Nov 11, 16:17 | #237 | 5GnpQR4…mkAQT | — | 0.855 | 0.008 | 121 |
| 26,344 | 9,483,862 | Nov 11, 15:15 | #237 | 5GnpQR4…mkAQT | — | 0.853 | 0.008 | 121 |
| 26,343 | 9,483,567 | Nov 11, 14:15 | #237 | 5GnpQR4…mkAQT | — | 0.853 | 0.008 | 121 |
| 26,342 | 9,483,103 | Nov 11, 12:43 | #237 | 5GnpQR4…mkAQT | — | 0.822 | 0.008 | 121 |
| 26,341 | 9,482,787 | Nov 11, 11:40 | #237 | 5GnpQR4…mkAQT | — | 0.822 | 0.008 | 121 |
| 26,340 | 9,482,356 | Nov 11, 10:14 | #237 | 5GnpQR4…mkAQT | — | 0.749 | 0.008 | 122 |
| 26,339 | 9,482,047 | Nov 11, 09:12 | #237 | 5GnpQR4…mkAQT | — | 0.819 | 0.008 | 120 |
| 26,338 | 9,481,739 | Nov 11, 08:10 | #237 | 5GnpQR4…mkAQT | — | 0.819 | 0.008 | 120 |
| 26,337 | 9,481,298 | Nov 11, 06:42 | #237 | 5GnpQR4…mkAQT | — | 0.830 | 0.008 | 120 |
| 26,336 | 9,480,999 | Nov 11, 05:42 | #237 | 5GnpQR4…mkAQT | — | 0.830 | 0.008 | 120 |
| 26,335 | 9,480,553 | Nov 11, 04:13 | #237 | 5GnpQR4…mkAQT | — | 0.856 | 0.009 | 116 |
| 26,334 | 9,480,280 | Nov 11, 03:18 | #237 | 5GnpQR4…mkAQT | — | 0.856 | 0.009 | 116 |
| 26,333 | 9,479,806 | Nov 11, 01:43 | #237 | 5GnpQR4…mkAQT | — | 0.838 | 0.009 | 117 |
| 26,332 | 9,479,492 | Nov 11, 00:41 | #237 | 5GnpQR4…mkAQT | — | 0.847 | 0.009 | 115 |
| Nov 10 | ||||||||
| 26,331 | 9,479,200 | Nov 10, 23:42 | #237 | 5GnpQR4…mkAQT | 5CkdyTX…KsPPF 0.01 | 0.847 | 0.009 | 115 |
| 26,330 | 9,478,795 | Nov 10, 22:21 | #237 | 5GnpQR4…mkAQT | 5CkdyTX…KsPPF 0.01 | 0.850 | 0.009 | 115 |
| 26,329 | 9,478,496 | Nov 10, 21:21 | #237 | 5GnpQR4…mkAQT | 5CkdyTX…KsPPF 0.01 | 0.850 | 0.009 | 115 |
| 26,328 | 9,478,026 | Nov 10, 19:47 | #237 | 5GnpQR4…mkAQT | 5CkdyTX…KsPPF 0.01 | 0.813 | 0.009 | 113 |
| 26,327 | 9,477,709 | Nov 10, 18:44 | #237 | 5GnpQR4…mkAQT | 5CkdyTX…KsPPF 0.01 | 0.797 | 0.009 | 113 |
| 26,326 | 9,477,420 | Nov 10, 17:46 | #237 | 5GnpQR4…mkAQT | 5CkdyTX…KsPPF 0.01 | 0.797 | 0.009 | 113 |
| 26,325 | 9,477,017 | Nov 10, 16:26 | #237 | 5GnpQR4…mkAQT | 5CkdyTX…KsPPF 0.01 | 0.827 | 0.010 | 99 |
| Nov 11 | ||||||||
| 26,324 | 9,476,918 | Nov 11, 06:56 | #237 | 5GnpQR4…mkAQT | — | 0.827 | 0.010 | 99 |
The chain does not store past Incentive vectors — history begins when taonets started watching (last 720 epochs kept), and every epoch is archived from live reads only. Gap rows mark stretches the worker never saw — they are downtime, not data. Winner coldkeys read at archival are exact; ones marked current were resolved later from the present ownership map. Race scores are the mission's own published scores for the settled round covering each epoch — retrospective (unknown until the round closed), in the mission's own units, never incentive.
who held the top incentive each day — 90 days back to 2026-08-13
| Day (UTC) | Winner uid | Winner coldkey | Top incentive | Block |
|---|---|---|---|---|
| 2026-11-10 | #237 | 5GnpQR4…mkAQT | 0.953 | 9,472,096 |
| 2026-11-09 | #237 | 5GnpQR4…mkAQT | 0.838 | 9,469,724 |
| 2026-11-08 | #237 | 5GnpQR4…mkAQT | 0.903 | 9,462,540 |
| 2026-11-07 | #237 | 5GnpQR4…mkAQT | 0.925 | 9,455,362 |
| 2026-11-06 | #237 | 5GnpQR4…mkAQT | 0.892 | 9,448,174 |
| 2026-11-05 | #237 | 5GnpQR4…mkAQT | 0.996 | 9,440,991 |
| 2026-11-04 | #237 | 5GnpQR4…mkAQT | 0.831 | 9,433,815 |
| 2026-11-03 | #237 | 5GnpQR4…mkAQT | 0.840 | 9,426,630 |
| 2026-11-02 | #237 | 5GnpQR4…mkAQT | 0.969 | 9,419,447 |
| 2026-11-01 | #237 | 5GnpQR4…mkAQT | 0.559 | 9,412,265 |
| 2026-10-31 | #237 | 5GnpQR4…mkAQT | 0.726 | 9,405,087 |
| 2026-10-30 | #237 | 5GnpQR4…mkAQT | 0.638 | 9,397,906 |
| 2026-10-29 | #237 | 5GnpQR4…mkAQT | 0.370 | 9,390,721 |
| 2026-10-28 | #237 | 5GnpQR4…mkAQT | 0.681 | 9,383,536 |
| 2026-10-27 | #237 | 5GnpQR4…mkAQT | 0.646 | 9,376,352 |
| 2026-10-26 | #237 | 5GnpQR4…mkAQT | 0.239 | 9,369,171 |
| 2026-10-25 | #237 | 5GnpQR4…mkAQT | 0.253 | 9,361,990 |
| 2026-10-24 | #237 | 5GnpQR4…mkAQT | 0.692 | 9,354,807 |
| 2026-10-23 | #237 | 5GnpQR4…mkAQT | 0.779 | 9,347,621 |
| 2026-10-22 | #237 | 5GnpQR4…mkAQT | 1.000 | 9,340,441 |
| 2026-10-21 | #237 | 5GnpQR4…mkAQT | 0.953 | 9,333,265 |
| 2026-10-20 | #237 | 5GnpQR4…mkAQT | 0.784 | 9,326,092 |
| 2026-10-19 | #237 | 5GnpQR4…mkAQT | 0.878 | 9,318,911 |
| 2026-10-18 | #237 | 5GnpQR4…mkAQT | 0.900 | 9,311,737 |
| 2026-10-17 | #237 | 5GnpQR4…mkAQT | 0.840 | 9,304,555 |
| 2026-10-16 | #237 | 5GnpQR4…mkAQT | 0.750 | 9,297,376 |
| 2026-10-15 | #237 | 5GnpQR4…mkAQT | 0.929 | 9,290,202 |
| 2026-10-14 | #237 | 5GnpQR4…mkAQT | 0.940 | 9,283,020 |
| 2026-10-13 | #237 | 5GnpQR4…mkAQT | 0.669 | 9,275,842 |
| 2026-10-12 | #237 | 5GnpQR4…mkAQT | 0.776 | 9,268,665 |
| 2026-10-11 | #237 | 5GnpQR4…mkAQT | 0.550 | 9,261,480 |
| 2026-10-10 | #237 | 5GnpQR4…mkAQT | 0.530 | 9,254,305 |
| 2026-10-09 | #237 | 5GnpQR4…mkAQT | 0.614 | 9,247,126 |
| 2026-10-08 | #237 | 5GnpQR4…mkAQT | 0.910 | 9,239,945 |
| 2026-10-07 | #237 | 5GnpQR4…mkAQT | 0.705 | 9,232,960 |
| 2026-10-06 | #237 | 5GnpQR4…mkAQT | 0.696 | 9,225,775 |
| 2026-10-05 | #237 | 5GnpQR4…mkAQT | 0.848 | 9,218,600 |
| 2026-10-04 | #237 | 5GnpQR4…mkAQT | 0.798 | 9,211,416 |
| 2026-10-03 | #237 | 5GnpQR4…mkAQT | 0.745 | 9,204,231 |
| 2026-10-02 | #237 | 5GnpQR4…mkAQT | 0.720 | 9,197,054 |
| 2026-10-01 | #237 | 5GnpQR4…mkAQT | 0.762 | 9,189,880 |
| 2026-09-30 | #237 | 5GnpQR4…mkAQT | 0.833 | 9,182,701 |
| 2026-09-29 | #237 | 5GnpQR4…mkAQT | 0.820 | 9,175,520 |
| 2026-09-28 | #237 | 5GnpQR4…mkAQT | 0.809 | 9,168,335 |
| 2026-09-27 | #237 | 5GnpQR4…mkAQT | 0.801 | 9,161,157 |
| 2026-09-26 | #237 | 5GnpQR4…mkAQT | 0.871 | 9,153,977 |
| 2026-09-25 | #237 | 5GnpQR4…mkAQT | 0.830 | 9,146,799 |
| 2026-09-24 | #237 | 5GnpQR4…mkAQT | 0.838 | 9,139,618 |
| 2026-09-23 | #237 | 5GnpQR4…mkAQT | 0.891 | 9,132,434 |
| 2026-09-22 | #237 | 5GnpQR4…mkAQT | 0.864 | 9,125,266 |
| 2026-09-21 | #237 | 5GnpQR4…mkAQT | 0.843 | 9,118,086 |
| 2026-09-20 | #237 | 5GnpQR4…mkAQT | 0.792 | 9,110,911 |
| 2026-09-19 | #237 | 5GnpQR4…mkAQT | 0.742 | 9,103,742 |
| 2026-09-18 | #237 | 5GnpQR4…mkAQT | 0.625 | 9,096,560 |
| 2026-09-17 | #237 | 5GnpQR4…mkAQT | 0.496 | 9,089,381 |
| 2026-09-16 | #237 | 5GnpQR4…mkAQT | 0.313 | 9,082,404 |
| 2026-09-15 | #237 | 5GnpQR4…mkAQT | 0.085 | 9,075,220 |
| 2026-09-14 | #24 | 5EP74DE…WP7o5 | 0.053 | 9,068,036 |
| 2026-09-13 | #51 | 5CQ6yxs…aWHNu | 0.050 | 9,060,857 |
| 2026-09-12 | #237 | 5GnpQR4…mkAQT | 0.043 | 9,053,670 |
| 2026-09-11 | #237 | 5GnpQR4…mkAQT | 0.168 | 9,046,491 |
| 2026-09-10 | #237 | 5GnpQR4…mkAQT | 0.348 | 9,039,306 |
| 2026-09-09 | #177 | 5GDyQJo…JKC3p | 0.048 | 9,032,126 |
| 2026-09-08 | #93 | 5Gxjw1p…1mf6e | 0.060 | 9,024,941 |
| 2026-09-07 | #93 | 5Gxjw1p…1mf6e | 0.048 | 9,017,756 |
| 2026-09-06 | #93 | 5Gxjw1p…1mf6e | 0.053 | 9,010,570 |
| 2026-09-05 | #93 | 5Gxjw1p…1mf6e | 0.059 | 9,003,384 |
| 2026-09-04 | #93 | 5Gxjw1p…1mf6e | 0.052 | 8,996,350 |
| 2026-09-03 | #93 | 5Gxjw1p…1mf6e | 0.050 | 8,989,170 |
| 2026-09-02 | #93 | 5Gxjw1p…1mf6e | 0.052 | 8,981,984 |
| 2026-09-01 | #24 | 5EP74DE…WP7o5 | 0.052 | 8,974,817 |
| 2026-08-31 | #24 | 5EP74DE…WP7o5 | 0.052 | 8,967,656 |
| 2026-08-30 | #210 | 5GjQrXY…ANe1X | 0.065 | 8,960,469 |
| 2026-08-29 | #239 | 5GjQrXY…ANe1X | 0.058 | 8,953,285 |
| 2026-08-28 | #172 | 5CkS5KP…BRF1L | 0.060 | 8,946,106 |
| 2026-08-27 | #172 | 5CkS5KP…BRF1L | 0.057 | 8,938,942 |
| 2026-08-26 | #228 | 5CkS5KP…BRF1L | 0.072 | 8,931,756 |
| 2026-08-25 | #172 | 5CkS5KP…BRF1L | 0.077 | 8,924,570 |
| 2026-08-24 | #228 | 5CkS5KP…BRF1L | 0.074 | 8,917,386 |
| 2026-08-23 | #170 | 5CQ6yxs…aWHNu | 0.057 | 8,910,201 |
| 2026-08-22 | #149 | 5CkS5KP…BRF1L | 0.052 | 8,903,016 |
| 2026-08-21 | #149 | 5CkS5KP…BRF1L | 0.054 | 8,895,835 |
| 2026-08-20 | #76 | 5D52i8f…duSkj | 0.050 | 8,888,665 |
| 2026-08-19 | #170 | 5CQ6yxs…aWHNu | 0.050 | 8,881,637 |
| 2026-08-18 | #0 | 5DUELqJ…7BtW5 | 0.522 | 8,874,456 |
| 2026-08-17 | #24 | 5EP74DE…WP7o5 | 0.133 | 8,867,276 |
| 2026-08-16 | #24 | 5EP74DE…WP7o5 | 0.098 | 8,852,907 |
| 2026-08-15 | #24 | 5EP74DE…WP7o5 | 0.094 | 8,845,726 |
| 2026-08-14 | #24 | 5EP74DE…WP7o5 | 0.089 | 8,838,546 |
| 2026-08-13 | #170 | 5CQ6yxs…aWHNu | 0.094 | 8,831,362 |
Sampled one block per day from archived chain state — the block number is the verifiable anchor, the date is ±drift. Identity (hotkey + coldkey) is read at the sample block, so it names who actually won that day. Days the endpoint had already pruned are absent, never zero-filled.
The Emission map — what the scores converted into, live.
No emissions flow on chain right now — the race pays nothing until that changes.
Zero τ is flowing to this subnet right now (chain reads taoInPerBlock = 0) — the shares above are the score distribution that would convert if flow resumes.
what it costs to race here, live
Best-case frame: every key wins every epoch. Your real edge depends on the model you bring — check the scoreboard and winner history above, and the repo watch before committing.