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 5GPT-5.4
Claude Sonnet 5

Anthropic

GPT-5.4

OpenAI

Pricing
Input $/1M$2.00$2.50
Output $/1M$10.00$15.00
Blended $/1M (3:1)$4.00$5.63
Cache read $/1M$0.20$0.25
Cache write $/1M$2.50
Quality indices
Intelligence Index53.451.4
Coding Index71.571.1
Math Index
Agentic Index46.741.1
Speed & latency
Output speed (tok/s)86149
Time to first token93.73s125.86s
Time to first answer token93.73s125.86s
Benchmarks (Artificial Analysis)
GPQA Diamond91.1%92.0%
Humanity's Last Exam39.6%41.6%
SciCode53.6%56.6%
IFBench73.9%
AA-LCR (Long Context Reasoning)70.7%74.0%
Terminal-Bench Hard57.6%
Terminal-Bench 2.180.5%78.3%
τ²-Bench (Telecom)87.1%
τ³-Bench Banking28.2%30.3%
Design Arena (Elo)
Website1,3141,247
Web apps1,3051,122
Full-stack1,2921,076
Mobile apps1,147
Android native1,2551,020
UI components1,3171,281
Data viz1,2811,268
SVG1,2391,241
3D1,3211,160
Game dev1,3451,298
Agentic game dev1,241
Godot game dev1,2571,135
HTML slides1,233
Python→PPTX slides1,252
Code categories1,3121,245
ASCII art1,2301,241
Specs
Context window1M1.05M
Max output tokens128K128K
Input modalitiestext, image, filetext, image, file
TokenizerClaudeGPT