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 5Gemini 3.1 Pro Preview
Claude Sonnet 5

Anthropic

Gemini 3.1 Pro Preview

Google

Pricing
Input $/1M$2.00$2.00
Output $/1M$10.00$12.00
Blended $/1M (3:1)$4.00$4.50
Cache read $/1M$0.20$0.20
Cache write $/1M$2.50$0.375
Quality indices
Intelligence Index53.446.5
Coding Index71.568.8
Math Index
Agentic Index46.721.4
Speed & latency
Output speed (tok/s)86135
Time to first token93.73s19.70s
Time to first answer token93.73s19.70s
Benchmarks (Artificial Analysis)
GPQA Diamond91.1%94.1%
Humanity's Last Exam39.6%44.7%
SciCode53.6%58.9%
IFBench77.1%
AA-LCR (Long Context Reasoning)70.7%72.7%
Terminal-Bench Hard53.8%
Terminal-Bench 2.180.5%73.8%
τ²-Bench (Telecom)95.6%
τ³-Bench Banking28.2%16.5%
Design Arena (Elo)
Website1,3141,278
Web apps1,3051,182
Full-stack1,2921,127
Mobile apps1,167
Android native1,2551,038
UI components1,3171,309
Data viz1,2811,262
SVG1,2391,334
3D1,3211,291
Game dev1,3451,258
Agentic game dev1,2411,128
Godot game dev1,2571,159
HTML slides1,2331,202
PPTX slides1,110
Python→PPTX slides1,2521,109
Agentic slides1,112
Agentic HTML slides1,226
Agentic slides (HTML)1,219
Agentic slides (Python PPTX)1,107
Code categories1,3121,274
ASCII art1,2301,312
Specs
Context window1M1.05M
Max output tokens128K66K
Input modalitiestext, image, fileaudio, file, image, text, video
TokenizerClaudeGemini