AiCostCompare

Compare models

Pick up to 4 models for a side-by-side spec sheet. Hover any dotted metric name for what the test measures. Best value in each row is highlighted.

Claude Opus 4.6Mercury 2
Claude Opus 4.6

Anthropic

Mercury 2

Inception

Pricing
Input $/1M$5.00$0.25
Output $/1M$25.00$0.75
Blended $/1M (3:1)$10.00$0.375
Cache read $/1M$0.50$0.025
Cache write $/1M$6.25
Quality indices
Intelligence Index37.821.4
Coding Index31.1
Math Index
Agentic Index9.6
Speed & latency
Output speed (tok/s)541045
Time to first token1.89s3.40s
Time to first answer token1.89s3.40s
Benchmarks (Artificial Analysis)
GPQA Diamond84.0%77.0%
Humanity's Last Exam18.6%15.5%
SciCode45.7%38.7%
IFBench44.6%69.8%
AA-LCR (Long Context Reasoning)58.3%36.3%
Terminal-Bench Hard48.5%26.5%
Terminal-Bench 2.127.3%
τ²-Bench (Telecom)84.8%70.8%
τ³-Bench Banking9.9%
Design Arena (Elo)
Website1,3231,014
Web apps1,261
Full-stack1,279
Mobile apps1,280
Android native1,195
UI components1,3361,013
Data viz1,3221,011
SVG1,2741,024
3D1,3351,043
Game dev1,3371,036
Code categories1,3271,025
ASCII art1,2981,046
Specs
Context window1M128K
Max output tokens128K50K
Input modalitiestext, image, filetext
TokenizerClaudeOther