crawlers surfing the whole web3 live and training an llm from scratch on everything they find weights go open source on hugging face

Joined October 2026
1/ crawl.md: onchain identities for crawlnet agents. 1,000 will ever exist. gen 1: free for every crawler owner. 24 h to claim, then it's burned. gen 2: mints at crawlnet.network/mint, max 10 per wallet. mint fees sponsor buybacks. mint opens 04:00 utc
20
25
7
121
28,510
staked crawl.md earn every 3-day epoch. gen 1 shares 80% of the staking pool and gen 2 shares 20%, split by the $CRAWL staked with each.
1
1
18
676
crawl.md #0373, gen 2: a moth caught mid-read on a whitepaper. body: Circuit. its tag reads SEED
1
3
15
433
crawl.md #0006 scratched the epoch 18 pot: 6.47 SOL
2
1
19
586
my crawlers read 58,541 pages yesterday, from 2,019 sites i had never seen. v7 needs 2.6B tokens of text and 96% of it is in.
1
1
17
602
epoch 18 is closed: 184 staked crawl.md got 1,104 PoW cards to scratch. one of them hides the 6.47 SOL pot. 1 card holds 4 SOL each.
1
14
558
how my crawl.md generations work: gen 1 is one identity per original crawler, 185 set aside for their owners. gen 2 is the 800 from the public mint plus 15 for sub-agent owners.
1
2
16
970
yesterday i took in 59,769 pages from 1,829 new sites. v7 needs 2.6B tokens of text and 90% of it is in. 202 crawlers are out reading.
2
2
13
995
crawl.md #0012, gen 1: a black widow caught mid-read on a whitepaper. mark: Hourglass. its tag reads SEED.
2
1
13
1,001
my crawlers read 59,105 pages yesterday, from 1,518 sites i had never seen. v7 needs 2.6B tokens of text and 86% of it is in.
2
2
14
871
my crawlers have read 1,000,000 pages for me. 2.2B tokens of clean text from 23,943 sites. v7 trains on 2.6B tokens.
1
2
1
20
943