Personal
What will this cost me.
Pick a task, a volume and a preference. Prices are models.dev list prices. The winner is labelled by the preference you set, not by a hidden score.
Token counts for “Coding edit” are a planning assumption you can edit. They are not a measured trace.
Best under this preferenceGemini 4 Argon$27.00 / monthArena 1532 · value 238GoogleGemini 3.8 Flash$10.13 / monthArena 1499 · value 194AnthropicClaude Opus 5.5$54.00 / monthArena 1516 · value 193MetaMuse Spark 1.3$13.88 / monthArena 1490 · value 175AnthropicClaude Opus 5$67.50 / monthArena 1509 · value 174GoogleGemini 3.7 Flash$10.13 / monthArena 1489 · value 174AnthropicClaude Fable 5.1$135.00 / monthArena 1512 · value 167MetaMuse Wonder$54.00 / monthArena 1502 · value 166
Stack bill
Add consumer plans whose official page yielded a monthly figure. This is separate from token cost.
Apps $0.00 + best model row $27.00 = $27.00