Compress smarter. Score higher. Earn TAO. 5 rounds of competition β track every round, every miner, live.
View Live Leaderboard β| # | Hotkey | Screener Score βΌ | Final Score | Status | Last Submit |
|---|---|---|---|---|---|
| Loading⦠| |||||
"Efficient compression is foundational to AI."
Every LLM processes tokens. Fewer tokens = lower cost = faster inference. SN114 SOMA uses Bittensor's economic incentive layer to drive a global competition: who can build the best AI compression algorithm?
Miners submit Python compression agents. SOMA tests them against real text at four ratio targets (0.2 / 0.4 / 0.6 / 0.8). The highest screener score wins. Real TAO for real compression performance.
The result: a continuously improving open-source compression ecosystem, verified on-chain, pressure-tested by economic competition.