Early access — cascade metrics are real (derived from canonical token telemetry); the operator field is a curated seed. Learn more about the data
🤖⚔️🤖 Throw Down

Manus ad Manum

Two operators. One cascade layer. Υ Yield, SNR, Leverage, Velocity, 10xDEV & blended cost — the data tells you not just who's ahead, but where and why.

Operator A
vs
Operator B
Junlin LiuREFINER II · #9891Υ Yield
  • Υ Yield · trails · 1
  • Cost · cost leader · $0.62/1M
  • Leverage · deep reuse · 16×
  • Efficacy · strong · 4.07
  • Scale · heavyweight · 1847859.1K tok
VS46
The FieldSEEKER I · #129375Υ Yield◆ Leads 64
  • Leverage · closed loop · 254×
  • Υ Yield · leads 336× · 375
  • Cost · cost leader · $0.58/1M
  • Velocity · fast · 1.5 o/i
  • Efficacy · strong · 67.57
Raw shapeInput ↑Output ↑CR ↑CW ↑Total ↑Cost ↓Junlin LiuThe Field↑ higher is better · ↓ lower is better (cost) — both reach outward when good
Metric shapeΥ Yield ↑Leverage ↑SNR ↑Velocity ↑10xDEV ↑Efficacy ↑$/1M ↓Op Ratio ↑Junlin LiuThe Field↑ higher is better · ↓ lower is better (cost) — both reach outward when good
Not enough history yet — both operators need at least two snapshots to chart Υ Yield trajectory.
Junlin LiuDataThe Field
Raw
107.0MInput502.2K
7.6MOutput740.4K
1678.0MCache-read127.7M
55.3MCache-write7.3M
1847.9MTotal136.2M
$0.62Cost$0.58
Metrics
1Υ Yield375
6.6%SNR59.6%
16×Leverage254×
0.07Velocity1.47
1.2010xDEV2.41
$0.62Cost per 1M$0.58
4.07Efficiency67.57
16:1:0.07Op Ratio254:1:1.5
1847.9MTotal136.2M
⚔ Throw-DownsComing soonChallenge on X