Writing Composite Index

BudgetOpenAI

Max

44.4score
$0.0030/ task
Best valueAnthropic

High

74.6score
$0.12/ task
Top scoreOpenAI

High

79.2score
$0.24/ task
Writing score versus cost Pareto frontierScoreCost / task (log)304050607080$0.003$0.005$0.008$0.02$0.03$0.05$0.08$0.2$0.3
Use Left, Right, Home, and End keys to compare qualifying models across spend caps, or Escape to clear.
ModelProviderScoreCost / taskInput / output ($/1M)
OpenAI79.2$0.24$8.00 / $40
Anthropic75.1$0.24$8.00 / $40
Anthropic74.6$0.12$4.00 / $20
Google67.8$0.060$2.00 / $10
OpenAI67.2$0.17$5.00 / $30
OpenAI66.3$0.12$4.00 / $20
OpenAI64.5$0.085$2.50 / $15
Meta64.1$0.12$4.00 / $20
Google60.0$0.045$1.50 / $7.50
Google58.1$0.022$0.75 / $3.75
OpenAI57.5$0.060$2.00 / $10
Meta56.9$0.029$1.25 / $4.25
SpaceXAI55.8$0.044$2.00 / $6.00
SpaceXAI54.3$0.022$1.25 / $2.50
Anthropic54.3$0.060$2.00 / $10
OpenAI53.6$0.0068$0.20 / $1.20
Google51.9$0.017$0.50 / $3.00
Google49.8$0.013$0.30 / $2.50
OpenAI44.4$0.0030$0.10 / $0.50
Anthropic40.2$0.030$1.00 / $5.00
Anthropic35.0$0.0030$0.10 / $0.50