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 5Qwen3.7 Max
Claude Sonnet 5

Anthropic

Qwen3.7 Max

Alibaba (Qwen)

Pricing
Input $/1M$2.00$1.48
Output $/1M$10.00$4.42
Blended $/1M (3:1)$4.00$2.21
Cache read $/1M$0.20$0.295
Cache write $/1M$2.50$1.84
Quality indices
Intelligence Index53.446
Coding Index71.566
Math Index
Agentic Index46.730.6
Speed & latency
Output speed (tok/s)86207
Time to first token93.73s1.55s
Time to first answer token93.73s13.16s
Benchmarks (Artificial Analysis)
GPQA Diamond91.1%92.3%
Humanity's Last Exam39.6%38.1%
SciCode53.6%48.8%
IFBench80.5%
AA-LCR (Long Context Reasoning)70.7%69.0%
Terminal-Bench Hard50.8%
Terminal-Bench 2.180.5%74.5%
τ²-Bench (Telecom)94.7%
τ³-Bench Banking28.2%10.9%
Design Arena (Elo)
Website1,3141,291
Web apps1,3051,253
Full-stack1,2921,225
Mobile apps1,208
Android native1,2551,182
UI components1,3171,313
Data viz1,2811,304
SVG1,2391,267
3D1,3211,310
Game dev1,3451,315
Agentic game dev1,2411,180
Godot game dev1,2571,248
HTML slides1,2331,191
Python→PPTX slides1,2521,229
Code categories1,3121,299
ASCII art1,2301,261
Specs
Context window1M1M
Max output tokens128K66K
Input modalitiestext, image, filetext
TokenizerClaudeQwen