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 4.6Qwen3.6 Plus
Claude Opus 4.6

Anthropic

Qwen3.6 Plus

Alibaba (Qwen)

Pricing
Input $/1M$5.00$0.325
Output $/1M$25.00$1.95
Blended $/1M (3:1)$10.00$0.731
Cache read $/1M$0.50
Cache write $/1M$6.25$0.406
Quality indices
Intelligence Index37.839.6
Coding Index54.5
Math Index
Agentic Index27.6
Speed & latency
Output speed (tok/s)5452
Time to first token1.89s1.73s
Time to first answer token1.89s107.62s
Benchmarks (Artificial Analysis)
GPQA Diamond84.0%88.2%
Humanity's Last Exam18.6%25.7%
SciCode45.7%40.7%
IFBench44.6%75.2%
AA-LCR (Long Context Reasoning)58.3%69.7%
Terminal-Bench Hard48.5%43.9%
Terminal-Bench 2.161.4%
τ²-Bench (Telecom)84.8%97.7%
τ³-Bench Banking16.5%
Design Arena (Elo)
Website1,3231,247
Web apps1,261
Full-stack1,279
Mobile apps1,280
Android native1,195
UI components1,3361,271
Data viz1,3221,255
SVG1,2741,207
3D1,3351,254
Game dev1,3371,262
Code categories1,3271,257
ASCII art1,2981,165
Specs
Context window1M1M
Max output tokens128K66K
Input modalitiestext, image, filetext, image, video
TokenizerClaudeQwen3