SIGNALAF

COMPARE

Operator A
vs
Operator B
4METRIC WINS6
#12 · 66.46RANK · Υ#1 · 976.9K
8.9M/dTHROUGHPUT88.2K/d

OPERATOR A LIVEGLOBAL #12
Lakshmi Narayanan Gglnarayanan
POWER II66Υ YIELD
  • Leverage · closed loop · 132×
  • Υ Yield · trails · 66
  • Cost · cost leader · $0.46/1M
  • Efficacy · strong · 34.05
  • Scale · heavyweight · 32577583.0K tok
Lakshmi Narayanan G · composite
first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataYield: 10 of 22 · 66.4643Leverage: 11 of 22 · 132.3844SNR: 11 of 22 · 0.3342Velocity: 11 of 22 · 0.502110xDEV: 11 of 22 · 2.1218Efficiency: 12 of 22 · 34.053Cost per 1M: 16 of 22 · 0.4571Yield: 10 of 22 points, value 66.4643Υ YieldYield: value 66.464310Leverage: 11 of 22 points, value 132.3844LeverageLeverage: value 132.384411SNR: 11 of 22 points, value 0.3342SNRSNR: value 0.334211Velocity: 11 of 22 points, value 0.5021VelocityVelocity: value 0.50211110xDEV: 11 of 22 points, value 2.121810xDEV10xDEV: value 2.121811Efficiency: 12 of 22 points, value 34.053EfficiencyEfficiency: value 34.05312Cost per 1M: 16 of 22 points, value 0.4571$/1MCost per 1M: value 0.457116

#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.

Lakshmi Narayanan G · composite: points among 22 live operators (1 to 22, higher is further out; cost per 1M inverted so cheaper is higher)
MeasureValuePoints
Yield66.464310
Leverage132.384411
SNR0.334211
Velocity0.502111
10xDEV2.121811
Efficiency34.05312
Cost per 1M0.457116
OPERATOR B LIVEGLOBAL #1
Karthikeyan KKkk_spartans
IGNITER I976.9KΥ YIELD
◆ LEADING · 6–4
  • Leverage · closed loop · 16.8K×
  • Υ Yield · leads 14.7K× · 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
Lakshmi Narayanan GVSKarthikeyan KK
4–6
HEAD-TO-HEAD · METRIC WINS
Lakshmi Narayanan G

Raw · Flow · Total

Karthikeyan KK
237.4MInput18.2K
119.2MOutput1.1M
31431.4MCache-read305.2M
789.5MCache-write15.8M
32577.6MTotal322.1M
$0.46Cost$0.52
8.9M/dThroughput88.2K/d
32577.6M∑ Total322.1M
Lakshmi Narayanan G

Metrics

Karthikeyan KK
66Υ Yield976.9K
33.4%SNR98.3%
132×Leverage16.8K×
0.50Velocity58.2
2.1210xDEV4.22
$0.46Cost per 1M$0.52
34.05Efficiency4428.17
132:1:0.50Op Ratio16786:1:58.2

Head to head

Cascade stages points in the field

Lakshmi Narayanan GKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataLakshmi Narayanan G: Transmission = 0.50 · 11 of 22Lakshmi Narayanan G: Commitment = 6.62 · 8 of 22Lakshmi Narayanan G: Reuse = 39.8 · 13 of 22Lakshmi Narayanan G: Leverage = 132 · 12 of 22Lakshmi Narayanan G: Υ Yield = 66.5 · 11 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 ÷ inputTransmissionLakshmi Narayanan G: Transmission = 0.50 · 11 of 22 points11Karthikeyan KK: Transmission = 58.2 · 22 of 22 points (more)● 22Commitment = cache write ÷ outputCommitmentLakshmi Narayanan G: Commitment = 6.62 · 8 of 22 points8Karthikeyan KK: Commitment = 14.9 · 20 of 22 points (more)● 20Reuse = cache read ÷ cache writeReuseLakshmi Narayanan G: Reuse = 39.8 · 13 of 22 points (more)● 13Karthikeyan KK: Reuse = 19.3 · 7 of 22 points7Leverage = T × C × R = cache read ÷ inputLeverageLakshmi Narayanan G: Leverage = 132 · 12 of 22 points12Karthikeyan KK: Leverage = 16.8K · 22 of 22 points (more)● 22Υ Yield = Leverage × Transmission = cache read × output ÷ input²Υ YieldLakshmi Narayanan G: Υ Yield = 66.5 · 11 of 22 points11Karthikeyan 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)
AxisFormulaLakshmi Narayanan GKarthikeyan KKMore
Transmissionoutput ÷ input0.50 (11)58.2 (22)Karthikeyan KK
Commitmentcache write ÷ output6.62 (8)14.9 (20)Karthikeyan KK
Reusecache read ÷ cache write39.8 (13)19.3 (7)Lakshmi Narayanan G
LeverageT × C × R = cache read ÷ input132 (12)16.8K (22)Karthikeyan KK
Υ YieldLeverage × Transmission = cache read × output ÷ input²66.5 (11)977K (22)Karthikeyan KK
Token mix share of own total

Lakshmi Narayanan GKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataLakshmi Narayanan G: Input ↓ = 0.73% · 12 of 22Lakshmi Narayanan G: Output = 0.37% · 9 of 22Lakshmi Narayanan G: Cache write = 2.4% · 9 of 22Lakshmi Narayanan G: Cache read = 96% · 16 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 ↓Lakshmi Narayanan G: Input ↓ = 0.73% · 12 of 22 points12Karthikeyan KK: Input ↓ = <0.01% · 22 of 22 points (more)● 22Output = output share of own totalOutputLakshmi Narayanan G: Output = 0.37% · 9 of 22 points (more)● 9Karthikeyan KK: Output = 0.33% · 5 of 22 points5Cache write = cache write share of own totalCache writeLakshmi Narayanan G: Cache write = 2.4% · 9 of 22 points9Karthikeyan KK: Cache write = 4.9% · 17 of 22 points (more)● 17Cache read = cache read share of own totalCache readLakshmi Narayanan G: Cache read = 96% · 16 of 22 points (more)● 16Karthikeyan 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)
AxisFormulaLakshmi Narayanan GKarthikeyan KKMore
Input ↓input share of own total (less is outward)0.73% (12)<0.01% (22)Karthikeyan KK
Outputoutput share of own total0.37% (9)0.33% (5)Lakshmi Narayanan G
Cache writecache write share of own total2.4% (9)4.9% (17)Karthikeyan KK
Cache readcache read share of own total96% (16)95% (13)Lakshmi Narayanan G
Υ YIELD · OVERTIMELakshmi Naraya…Karthikeyan KK0.0334.7K669.3K1.0M1.3M66.5976.9K07-2607-31
Υ YIELD · OVERTIMELakshmi Nar…Karthikeyan…0.0669.3K1.3M66.5976.9K07-2607-31

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.