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 Sonnet 5Qwen3 Max
Claude Sonnet 5

Anthropic

Qwen3 Max

Alibaba (Qwen)

Pricing
Input $/1M$2.00$0.78
Output $/1M$10.00$3.90
Blended $/1M (3:1)$4.00$1.56
Cache read $/1M$0.20$0.156
Cache write $/1M$2.50$0.975
Quality indices
Intelligence Index53.424
Coding Index71.5
Math Index80.7
Agentic Index46.7
Speed & latency
Output speed (tok/s)8662
Time to first token93.73s1.77s
Time to first answer token93.73s1.77s
Benchmarks (Artificial Analysis)
MMLU-Pro84.1%
GPQA Diamond91.1%76.4%
Humanity's Last Exam39.6%11.1%
LiveCodeBench76.7%
SciCode53.6%38.3%
AIME 202580.7%
IFBench44.1%
AA-LCR (Long Context Reasoning)70.7%46.7%
Terminal-Bench Hard20.5%
Terminal-Bench 2.180.5%
τ²-Bench (Telecom)74.3%
τ³-Bench Banking28.2%
Design Arena (Elo)
Website1,3141,146
Web apps1,305
Full-stack1,292
Android native1,255
UI components1,3171,119
Data viz1,2811,141
SVG1,2391,058
3D1,3211,132
Game dev1,3451,153
Agentic game dev1,241
Godot game dev1,257
HTML slides1,233
Python→PPTX slides1,252
Code categories1,3121,143
ASCII art1,2301,170
Specs
Context window1M262K
Max output tokens128K33K
Input modalitiestext, image, filetext
TokenizerClaudeQwen3