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

COMPARE

Operator A
vs
Operator B
3METRIC WINS7
#6 · 19.1KRANK · Υ#1 · 976.9K
11.6M/dTHROUGHPUT88.2K/d

OPERATOR A LIVEGLOBAL #6
Anagh Kumar Baranwalsignal-148d593199
POWER I19.1KΥ YIELD
  • Leverage · closed loop · 1.7K×
  • Υ Yield · trails · 19.1K
  • Cost · cost leader · $0.57/1M
  • Velocity · fast · 11.4 o/i
  • Efficacy · strong · 441.87
Anagh Kumar Baranwal · composite
first: #1 Karthikeyan KKmiddle: #11 Nikita Reshetniklast: #22 DirtyDataYield: 16 of 22 · 19,081.9399Leverage: 15 of 22 · 1,667.2722SNR: 19 of 22 · 0.9196Velocity: 19 of 22 · 11.44510xDEV: 15 of 22 · 3.222Efficiency: 16 of 22 · 441.8695Cost per 1M: 10 of 22 · 0.5698Yield: 16 of 22 points, value 19,081.9399Υ YieldYield: value 19,081.939916Leverage: 15 of 22 points, value 1,667.2722LeverageLeverage: value 1,667.272215SNR: 19 of 22 points, value 0.9196SNRSNR: value 0.919619Velocity: 19 of 22 points, value 11.445VelocityVelocity: value 11.4451910xDEV: 15 of 22 points, value 3.22210xDEV10xDEV: value 3.22215Efficiency: 16 of 22 points, value 441.8695EfficiencyEfficiency: value 441.869516Cost per 1M: 10 of 22 points, value 0.5698$/1MCost per 1M: value 0.569810

#1 Karthikeyan KK#11 Nikita Reshetnik#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.

Anagh Kumar Baranwal · composite: points among 22 live operators (1 to 22, higher is further out; cost per 1M inverted so cheaper is higher)
MeasureValuePoints
Yield19,081.939916
Leverage1,667.272215
SNR0.919619
Velocity11.44519
10xDEV3.22215
Efficiency441.869516
Cost per 1M0.569810
OPERATOR B LIVEGLOBAL #1
Karthikeyan KKkk_spartans
IGNITER I976.9KΥ YIELD
◆ LEADING · 7–3
  • Leverage · closed loop · 16.8K×
  • Υ Yield · leads 51× · 976.9K
  • Cost · cost leader · $0.52/1M
  • Velocity · fast · 58.2 o/i
  • Efficacy · strong · 4428.17
Karthikeyan KK · composite
first: #1 Karthikeyan KKmiddle: #11 Nikita Reshetniklast: #22 DirtyDataYield: 21 of 22 · 976,944.2293Leverage: 21 of 22 · 16,785.7105SNR: 22 of 22 · 0.9831Velocity: 22 of 22 · 58.200910xDEV: 21 of 22 · 4.2249Efficiency: 22 of 22 · 4,428.1743Cost per 1M: 14 of 22 · 0.5177Yield: 21 of 22 points, value 976,944.2293Υ YieldYield: value 976,944.229321Leverage: 21 of 22 points, value 16,785.7105LeverageLeverage: value 16,785.710521SNR: 22 of 22 points, value 0.9831SNRSNR: value 0.983122Velocity: 22 of 22 points, value 58.2009VelocityVelocity: value 58.20092210xDEV: 21 of 22 points, value 4.224910xDEV10xDEV: value 4.224921Efficiency: 22 of 22 points, value 4,428.1743EfficiencyEfficiency: value 4,428.174322Cost per 1M: 14 of 22 points, value 0.5177$/1MCost per 1M: value 0.517714

#1 Karthikeyan KK#11 Nikita Reshetnik#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
Anagh Kumar BaranwalVSKarthikeyan KK
3–7
HEAD-TO-HEAD · METRIC WINS
Anagh Kumar Baranwal

Raw · Flow · Total

Karthikeyan KK
23.9MInput18.2K
273.3MOutput1.1M
39820.0MCache-read305.2M
2119.9MCache-write15.8M
42237.1MTotal322.1M
$0.57Cost$0.52
11.6M/dThroughput88.2K/d
42237.1M∑ Total322.1M
Anagh Kumar Baranwal

Metrics

Karthikeyan KK
19.1KΥ Yield976.9K
92.0%SNR98.3%
1.7K×Leverage16.8K×
11.4Velocity58.2
3.2210xDEV4.22
$0.57Cost per 1M$0.52
441.87Efficiency4428.17
1667:1:11.4Op Ratio16786:1:58.2

Head to head

Cascade stages points in the field

Anagh Kumar BaranwalKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Nikita Reshetniklast: #22 DirtyDataAnagh Kumar Baranwal: Transmission = 11.4 · 19 of 22Anagh Kumar Baranwal: Commitment = 7.76 · 12 of 22Anagh Kumar Baranwal: Reuse = 18.8 · 6 of 22Anagh Kumar Baranwal: Leverage = 1.7K · 16 of 22Anagh Kumar Baranwal: Υ Yield = 19.1K · 17 of 22Karthikeyan KK: Transmission = 58.2 · 22 of 22Karthikeyan KK: Commitment = 14.9 · 19 of 22Karthikeyan KK: Reuse = 19.3 · 8 of 22Karthikeyan KK: Leverage = 16.8K · 22 of 22Karthikeyan KK: Υ Yield = 977K · 22 of 22Transmission = output ÷ inputTransmissionAnagh Kumar Baranwal: Transmission = 11.4 · 19 of 22 points19Karthikeyan KK: Transmission = 58.2 · 22 of 22 points (more)● 22Commitment = cache write ÷ outputCommitmentAnagh Kumar Baranwal: Commitment = 7.76 · 12 of 22 points12Karthikeyan KK: Commitment = 14.9 · 19 of 22 points (more)● 19Reuse = cache read ÷ cache writeReuseAnagh Kumar Baranwal: Reuse = 18.8 · 6 of 22 points6Karthikeyan KK: Reuse = 19.3 · 8 of 22 points (more)● 8Leverage = T × C × R = cache read ÷ inputLeverageAnagh Kumar Baranwal: Leverage = 1.7K · 16 of 22 points16Karthikeyan KK: Leverage = 16.8K · 22 of 22 points (more)● 22Υ Yield = Leverage × Transmission = cache read × output ÷ input²Υ YieldAnagh Kumar Baranwal: Υ Yield = 19.1K · 17 of 22 points17Karthikeyan KK: Υ Yield = 977K · 22 of 22 points (more)● 22

#1 Karthikeyan KK#11 Nikita Reshetnik#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)
AxisFormulaAnagh Kumar BaranwalKarthikeyan KKMore
Transmissionoutput ÷ input11.4 (19)58.2 (22)Karthikeyan KK
Commitmentcache write ÷ output7.76 (12)14.9 (19)Karthikeyan KK
Reusecache read ÷ cache write18.8 (6)19.3 (8)Karthikeyan KK
LeverageT × C × R = cache read ÷ input1.7K (16)16.8K (22)Karthikeyan KK
Υ YieldLeverage × Transmission = cache read × output ÷ input²19.1K (17)977K (22)Karthikeyan KK
Token mix share of own total

Anagh Kumar BaranwalKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Nikita Reshetniklast: #22 DirtyDataAnagh Kumar Baranwal: Input ↓ = 0.06% · 16 of 22Anagh Kumar Baranwal: Output = 0.65% · 17 of 22Anagh Kumar Baranwal: Cache write = 5.0% · 18 of 22Anagh Kumar Baranwal: Cache read = 94% · 10 of 22Karthikeyan KK: Input ↓ = <0.01% · 22 of 22Karthikeyan KK: Output = 0.33% · 4 of 22Karthikeyan KK: Cache write = 4.9% · 16 of 22Karthikeyan KK: Cache read = 95% · 13 of 22Input ↓ = input share of own total (less is outward)Input ↓Anagh Kumar Baranwal: Input ↓ = 0.06% · 16 of 22 points16Karthikeyan KK: Input ↓ = <0.01% · 22 of 22 points (more)● 22Output = output share of own totalOutputAnagh Kumar Baranwal: Output = 0.65% · 17 of 22 points (more)● 17Karthikeyan KK: Output = 0.33% · 4 of 22 points4Cache write = cache write share of own totalCache writeAnagh Kumar Baranwal: Cache write = 5.0% · 18 of 22 points (more)● 18Karthikeyan KK: Cache write = 4.9% · 16 of 22 points16Cache read = cache read share of own totalCache readAnagh Kumar Baranwal: Cache read = 94% · 10 of 22 points10Karthikeyan KK: Cache read = 95% · 13 of 22 points (more)● 13

#1 Karthikeyan KK#11 Nikita Reshetnik#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)
AxisFormulaAnagh Kumar BaranwalKarthikeyan KKMore
Input ↓input share of own total (less is outward)0.06% (16)<0.01% (22)Karthikeyan KK
Outputoutput share of own total0.65% (17)0.33% (4)Anagh Kumar Baranwal
Cache writecache write share of own total5.0% (18)4.9% (16)Anagh Kumar Baranwal
Cache readcache read share of own total94% (10)95% (13)Karthikeyan KK
Υ YIELD · OVERTIMEAnagh Kumar Ba…Karthikeyan KK0.0334.7K669.3K1.0M1.3M19.1K976.9K07-1407-1507-1707-1807-2207-2608-03
Υ YIELD · OVERTIMEAnagh Kumar…Karthikeyan…0.0669.3K1.3M19.1K976.9K07-1407-1707-2208-03

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.