SIGNALAF

COMPARE

Operator A
vs
Operator B
4METRIC WINS6
#4 · 34.8KRANK · Υ#1 · 976.9K
13.6M/dTHROUGHPUT88.2K/d

OPERATOR A LIVEGLOBAL #4
Kabir AcharyaKabir4charya
POWER I34.8KΥ YIELD
  • Leverage · closed loop · 3.1K×
  • Υ Yield · trails · 34.8K
  • Cost · cost leader · $0.41/1M
  • Velocity · fast · 11.1 o/i
  • Efficacy · strong · 802.47
Kabir Acharya · composite
first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataYield: 18 of 22 · 34,768.6661Leverage: 18 of 22 · 3,144.0353SNR: 18 of 22 · 0.9171Velocity: 18 of 22 · 11.058610xDEV: 18 of 22 · 3.4975Efficiency: 19 of 22 · 802.4684Cost per 1M: 20 of 22 · 0.4103Yield: 18 of 22 points, value 34,768.6661Υ YieldYield: value 34,768.666118Leverage: 18 of 22 points, value 3,144.0353LeverageLeverage: value 3,144.035318SNR: 18 of 22 points, value 0.9171SNRSNR: value 0.917118Velocity: 18 of 22 points, value 11.0586VelocityVelocity: value 11.05861810xDEV: 18 of 22 points, value 3.497510xDEV10xDEV: value 3.497518Efficiency: 19 of 22 points, value 802.4684EfficiencyEfficiency: value 802.468419Cost per 1M: 20 of 22 points, value 0.4103$/1MCost per 1M: value 0.410320

#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.

Kabir Acharya · composite: points among 22 live operators (1 to 22, higher is further out; cost per 1M inverted so cheaper is higher)
MeasureValuePoints
Yield34,768.666118
Leverage3,144.035318
SNR0.917118
Velocity11.058618
10xDEV3.497518
Efficiency802.468419
Cost per 1M0.410320
OPERATOR B LIVEGLOBAL #1
Karthikeyan KKkk_spartans
IGNITER I976.9KΥ YIELD
◆ LEADING · 6–4
  • Leverage · closed loop · 16.8K×
  • Υ Yield · leads 28× · 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
Kabir AcharyaVSKarthikeyan KK
4–6
HEAD-TO-HEAD · METRIC WINS
Kabir Acharya

Raw · Flow · Total

Karthikeyan KK
15.4MInput18.2K
170.4MOutput1.1M
48454.9MCache-read305.2M
844.2MCache-write15.8M
49485.0MTotal322.1M
$0.41Cost$0.52
13.6M/dThroughput88.2K/d
49485.0M∑ Total322.1M
Kabir Acharya

Metrics

Karthikeyan KK
34.8KΥ Yield976.9K
91.7%SNR98.3%
3.1K×Leverage16.8K×
11.1Velocity58.2
3.5010xDEV4.22
$0.41Cost per 1M$0.52
802.47Efficiency4428.17
3144:1:11.1Op Ratio16786:1:58.2

Head to head

Cascade stages points in the field

Kabir AcharyaKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataKabir Acharya: Transmission = 11.1 · 18 of 22Kabir Acharya: Commitment = 4.95 · 5 of 22Kabir Acharya: Reuse = 57.4 · 18 of 22Kabir Acharya: Leverage = 3.1K · 19 of 22Kabir Acharya: Υ Yield = 34.8K · 19 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 ÷ inputTransmissionKabir Acharya: Transmission = 11.1 · 18 of 22 points18Karthikeyan KK: Transmission = 58.2 · 22 of 22 points (more)● 22Commitment = cache write ÷ outputCommitmentKabir Acharya: Commitment = 4.95 · 5 of 22 points5Karthikeyan KK: Commitment = 14.9 · 20 of 22 points (more)● 20Reuse = cache read ÷ cache writeReuseKabir Acharya: Reuse = 57.4 · 18 of 22 points (more)● 18Karthikeyan KK: Reuse = 19.3 · 7 of 22 points7Leverage = T × C × R = cache read ÷ inputLeverageKabir Acharya: Leverage = 3.1K · 19 of 22 points19Karthikeyan KK: Leverage = 16.8K · 22 of 22 points (more)● 22Υ Yield = Leverage × Transmission = cache read × output ÷ input²Υ YieldKabir Acharya: Υ Yield = 34.8K · 19 of 22 points19Karthikeyan 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)
AxisFormulaKabir AcharyaKarthikeyan KKMore
Transmissionoutput ÷ input11.1 (18)58.2 (22)Karthikeyan KK
Commitmentcache write ÷ output4.95 (5)14.9 (20)Karthikeyan KK
Reusecache read ÷ cache write57.4 (18)19.3 (7)Kabir Acharya
LeverageT × C × R = cache read ÷ input3.1K (19)16.8K (22)Karthikeyan KK
Υ YieldLeverage × Transmission = cache read × output ÷ input²34.8K (19)977K (22)Karthikeyan KK
Token mix share of own total

Kabir AcharyaKarthikeyan KK

first: #1 Karthikeyan KKmiddle: #11 Georgelast: #22 DirtyDataKabir Acharya: Input ↓ = 0.03% · 19 of 22Kabir Acharya: Output = 0.34% · 6 of 22Kabir Acharya: Cache write = 1.7% · 5 of 22Kabir Acharya: Cache read = 98% · 21 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 ↓Kabir Acharya: Input ↓ = 0.03% · 19 of 22 points19Karthikeyan KK: Input ↓ = <0.01% · 22 of 22 points (more)● 22Output = output share of own totalOutputKabir Acharya: Output = 0.34% · 6 of 22 points (more)● 6Karthikeyan KK: Output = 0.33% · 5 of 22 points5Cache write = cache write share of own totalCache writeKabir Acharya: Cache write = 1.7% · 5 of 22 points5Karthikeyan KK: Cache write = 4.9% · 17 of 22 points (more)● 17Cache read = cache read share of own totalCache readKabir Acharya: Cache read = 98% · 21 of 22 points (more)● 21Karthikeyan 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)
AxisFormulaKabir AcharyaKarthikeyan KKMore
Input ↓input share of own total (less is outward)0.03% (19)<0.01% (22)Karthikeyan KK
Outputoutput share of own total0.34% (6)0.33% (5)Kabir Acharya
Cache writecache write share of own total1.7% (5)4.9% (17)Karthikeyan KK
Cache readcache read share of own total98% (21)95% (13)Kabir Acharya
Υ YIELD · OVERTIMEKabir AcharyaKarthikeyan KK0.038.5M77.0M115.4M153.9M34.8K976.9K07-1407-1707-2610-0210-1110-24
Υ YIELD · OVERTIMEKabir Achar…Karthikeyan…0.077.0M153.9M34.8K976.9K07-1407-2610-1110-24

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.