SIGNALAF

COMPARE

Operator A
vs
Operator B
3METRIC WINS7
#14 · 46.06RANK · Υ#1 · 976.9K
2.1M/dTHROUGHPUT88.2K/d

OPERATOR A LIVEGLOBAL #14
kr-yeonsignal-51c11cb75e
SEEKER I46Υ YIELD
  • Υ Yield · trails · 46
  • Cost · cost leader · $0.52/1M
  • Leverage · deep reuse · 92×
  • Efficacy · strong · 24.01
  • Scale · heavyweight · 7652493.6K tok
kr-yeon · composite
first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataYield: 8 of 22 · 46.057Leverage: 9 of 22 · 92.114SNR: 9 of 22 · 0.3333Velocity: 9 of 22 · 0.510xDEV: 9 of 22 · 1.9643Efficiency: 9 of 22 · 24.0052Cost per 1M: 11 of 22 · 0.5247Yield: 8 of 22 points, value 46.057Υ YieldYield: value 46.0578Leverage: 9 of 22 points, value 92.114LeverageLeverage: value 92.1149SNR: 9 of 22 points, value 0.3333SNRSNR: value 0.33339Velocity: 9 of 22 points, value 0.5VelocityVelocity: value 0.5910xDEV: 9 of 22 points, value 1.964310xDEV10xDEV: value 1.96439Efficiency: 9 of 22 points, value 24.0052EfficiencyEfficiency: value 24.00529Cost per 1M: 11 of 22 points, value 0.5247$/1MCost per 1M: value 0.524711

#1 Karthikeyan KK#11 George#22 DirtyData

Points among 22 live operators (1 to 22); outward = more. Cost per 1M: cheaper is outward. Token-flow shape, not a quality score.

kr-yeon · composite: points among 22 live operators (1 to 22, higher is further out; cost per 1M inverted so cheaper is higher)
MeasureValuePoints
Yield46.0578
Leverage92.1149
SNR0.33339
Velocity0.59
10xDEV1.96439
Efficiency24.00529
Cost per 1M0.524711
OPERATOR B LIVEGLOBAL #1
Karthikeyan KKkk_spartans
IGNITER I976.9KΥ YIELD
◆ LEADING · 7–3
  • Leverage · closed loop · 16.8K×
  • Υ Yield · leads 21.2K× · 976.9K
  • Cost · cost leader · $0.52/1M
  • Velocity · fast · 58.2 o/i
  • Efficacy · strong · 4428.17
Karthikeyan KK · composite
first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataYield: 21 of 22 · 976,944.2293Leverage: 21 of 22 · 16,785.7105SNR: 22 of 22 · 0.9831Velocity: 22 of 22 · 58.200910xDEV: 21 of 22 · 4.2249Efficiency: 22 of 22 · 4,428.1743Cost per 1M: 14 of 22 · 0.5177Yield: 21 of 22 points, value 976,944.2293Υ YieldYield: value 976,944.229321Leverage: 21 of 22 points, value 16,785.7105LeverageLeverage: value 16,785.710521SNR: 22 of 22 points, value 0.9831SNRSNR: value 0.983122Velocity: 22 of 22 points, value 58.2009VelocityVelocity: value 58.20092210xDEV: 21 of 22 points, value 4.224910xDEV10xDEV: value 4.224921Efficiency: 22 of 22 points, value 4,428.1743EfficiencyEfficiency: value 4,428.174322Cost per 1M: 14 of 22 points, value 0.5177$/1MCost per 1M: value 0.517714

#1 Karthikeyan KK#11 George#22 DirtyData

Points among 22 live operators (1 to 22); outward = more. Cost per 1M: cheaper is outward. Token-flow shape, not a quality score.

Karthikeyan KK · composite: points among 22 live operators (1 to 22, higher is further out; cost per 1M inverted so cheaper is higher)
MeasureValuePoints
Yield976,944.229321
Leverage16,785.710521
SNR0.983122
Velocity58.200922
10xDEV4.224921
Efficiency4,428.174322
Cost per 1M0.517714
MATCHUP · BAR STATISTICS
kr-yeonVSKarthikeyan KK
3–7
HEAD-TO-HEAD · METRIC WINS
kr-yeon

Raw · Flow · Total

Karthikeyan KK
78.9MInput18.2K
39.4MOutput1.1M
7265.5MCache-read305.2M
268.7MCache-write15.8M
7652.5MTotal322.1M
$0.52Cost$0.52
2.1M/dThroughput88.2K/d
7652.5M∑ Total322.1M
kr-yeon

Metrics

Karthikeyan KK
46Υ Yield976.9K
33.3%SNR98.3%
92×Leverage16.8K×
0.50Velocity58.2
1.9610xDEV4.22
$0.52Cost per 1M$0.52
24.01Efficiency4428.17
92:1:0.50Op Ratio16786:1:58.2

Head to head

Cascade stages points in the field

kr-yeonKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDatakr-yeon: Transmission = 0.50 · 9 of 22kr-yeon: Commitment = 6.81 · 9 of 22kr-yeon: Reuse = 27.0 · 11 of 22kr-yeon: Leverage = 92.1 · 10 of 22kr-yeon: Υ Yield = 46.1 · 9 of 22Karthikeyan KK: Transmission = 58.2 · 22 of 22Karthikeyan KK: Commitment = 14.9 · 20 of 22Karthikeyan KK: Reuse = 19.3 · 7 of 22Karthikeyan KK: Leverage = 16.8K · 22 of 22Karthikeyan KK: Υ Yield = 977K · 22 of 22Transmission = output ÷ inputTransmissionkr-yeon: Transmission = 0.50 · 9 of 22 points9Karthikeyan KK: Transmission = 58.2 · 22 of 22 points (more)● 22Commitment = cache write ÷ outputCommitmentkr-yeon: Commitment = 6.81 · 9 of 22 points9Karthikeyan KK: Commitment = 14.9 · 20 of 22 points (more)● 20Reuse = cache read ÷ cache writeReusekr-yeon: Reuse = 27.0 · 11 of 22 points (more)● 11Karthikeyan KK: Reuse = 19.3 · 7 of 22 points7Leverage = T × C × R = cache read ÷ inputLeveragekr-yeon: Leverage = 92.1 · 10 of 22 points10Karthikeyan KK: Leverage = 16.8K · 22 of 22 points (more)● 22Υ Yield = Leverage × Transmission = cache read × output ÷ input²Υ Yieldkr-yeon: Υ Yield = 46.1 · 9 of 22 points9Karthikeyan KK: Υ Yield = 977K · 22 of 22 points (more)● 22

#1 Karthikeyan KK#11 George#22 DirtyData

Where a Yield gap comes from. Transmission × Commitment × Reuse = Leverage (cache read ÷ input); Υ = Leverage × Transmission. Dot = more points on that axis.

Cascade stages: points among 22 live operators (1 to 22)
AxisFormulakr-yeonKarthikeyan KKMore
Transmissionoutput ÷ input0.50 (9)58.2 (22)Karthikeyan KK
Commitmentcache write ÷ output6.81 (9)14.9 (20)Karthikeyan KK
Reusecache read ÷ cache write27.0 (11)19.3 (7)kr-yeon
LeverageT × C × R = cache read ÷ input92.1 (10)16.8K (22)Karthikeyan KK
Υ YieldLeverage × Transmission = cache read × output ÷ input²46.1 (9)977K (22)Karthikeyan KK
Token mix share of own total

kr-yeonKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDatakr-yeon: Input ↓ = 1.0% · 9 of 22kr-yeon: Output = 0.52% · 14 of 22kr-yeon: Cache write = 3.5% · 13 of 22kr-yeon: Cache read = 95% · 14 of 22Karthikeyan KK: Input ↓ = <0.01% · 22 of 22Karthikeyan KK: Output = 0.33% · 5 of 22Karthikeyan KK: Cache write = 4.9% · 17 of 22Karthikeyan KK: Cache read = 95% · 13 of 22Input ↓ = input share of own total (less is outward)Input ↓kr-yeon: Input ↓ = 1.0% · 9 of 22 points9Karthikeyan KK: Input ↓ = <0.01% · 22 of 22 points (more)● 22Output = output share of own totalOutputkr-yeon: Output = 0.52% · 14 of 22 points (more)● 14Karthikeyan KK: Output = 0.33% · 5 of 22 points5Cache write = cache write share of own totalCache writekr-yeon: Cache write = 3.5% · 13 of 22 points13Karthikeyan KK: Cache write = 4.9% · 17 of 22 points (more)● 17Cache read = cache read share of own totalCache readkr-yeon: Cache read = 95% · 14 of 22 points (more)● 14Karthikeyan KK: Cache read = 95% · 13 of 22 points13

#1 Karthikeyan KK#11 George#22 DirtyData

How each operates, not how much: each pillar's share of the operator's own tokens, ranked against the field. Input share is flipped, so less input is outward.

Token mix: points among 22 live operators (1 to 22)
AxisFormulakr-yeonKarthikeyan KKMore
Input ↓input share of own total (less is outward)1.0% (9)<0.01% (22)Karthikeyan KK
Outputoutput share of own total0.52% (14)0.33% (5)kr-yeon
Cache writecache write share of own total3.5% (13)4.9% (17)Karthikeyan KK
Cache readcache read share of own total95% (14)95% (13)kr-yeon
Υ YIELD · OVERTIMEkr-yeonKarthikeyan KK0.0334.7K669.3K1.0M1.3M46.1976.9K07-1507-1607-1807-2107-26
Υ YIELD · OVERTIMEkr-yeonKarthikeyan…0.0669.3K1.3M46.1976.9K07-1507-1807-26

FAQ · 3 questions

How do I compare two AI operators on SigRank?

Visit signalaf.com/compare and select two operators using the dropdown selectors. The comparison shows Yield (Υ), SNR, Leverage, Velocity, 10xDEV, and blended cost side by side, plus each operator's own raw and metric radars and a shared Yield-over-time chart. You can share the comparison via URL.

What metrics does the SigRank comparison show?

The head-to-head comparison shows six cascade metrics: Yield (Υ = cache_read × output / input²), SNR (signal-to-noise ratio), Leverage (cache reuse efficiency), Velocity (output per token), 10xDEV (log-normalized value above replacement), and blended cost ($/1M tokens). Individual raw and metric radar charts describe each live operator, with their recorded Yield history beneath.

Can I compare myself against another AI operator?

Yes. If you are enrolled on SigRank, the Compare Against Me feature lets you select yourself as one of the comparison operands. You can compare your cascade metrics against any other operator on the leaderboard to see exactly where and why you differ.