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 4.6Claude Sonnet 5
Claude Opus 4.6

Anthropic

Claude Sonnet 5

Anthropic

Pricing
Input $/1M$5.00$2.00
Output $/1M$25.00$10.00
Blended $/1M (3:1)$10.00$4.00
Cache read $/1M$0.50$0.20
Cache write $/1M$6.25$2.50
Quality indices
Intelligence Index37.853.4
Coding Index71.5
Math Index
Agentic Index46.7
Speed & latency
Output speed (tok/s)5486
Time to first token1.89s93.73s
Time to first answer token1.89s93.73s
Benchmarks (Artificial Analysis)
GPQA Diamond84.0%91.1%
Humanity's Last Exam18.6%39.6%
SciCode45.7%53.6%
IFBench44.6%
AA-LCR (Long Context Reasoning)58.3%70.7%
Terminal-Bench Hard48.5%
Terminal-Bench 2.180.5%
τ²-Bench (Telecom)84.8%
τ³-Bench Banking28.2%
Design Arena (Elo)
Website1,3231,314
Web apps1,2611,305
Full-stack1,2791,292
Mobile apps1,280
Android native1,1951,255
UI components1,3361,317
Data viz1,3221,281
SVG1,2741,239
3D1,3351,321
Game dev1,3371,345
Agentic game dev1,241
Godot game dev1,257
HTML slides1,233
Python→PPTX slides1,252
Code categories1,3271,312
ASCII art1,2981,230
Specs
Context window1M1M
Max output tokens128K128K
Input modalitiestext, image, filetext, image, file
TokenizerClaudeClaude