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 4.1 (batch)GPT-4.1
Claude Opus 4.1 (batch)

Anthropic

GPT-4.1

OpenAI

Pricing
Input $/1M$7.50$2.00
Output $/1M$37.50$8.00
Blended $/1M (3:1)$15.00$3.50
Cache read $/1M$0.75$0.50
Cache write $/1M$9.38
Quality indices
Intelligence Index12.7
Coding Index
Math Index34.7
Agentic Index
Speed & latency
Output speed (tok/s)0
Time to first token0s
Time to first answer token0s
Benchmarks (Artificial Analysis)
MMLU-Pro80.6%
GPQA Diamond66.6%
Humanity's Last Exam4.2%
LiveCodeBench45.7%
MATH-50091.3%
AIME43.7%
AIME 202534.7%
IFBench43.0%
AA-LCR (Long Context Reasoning)68.3%
Terminal-Bench Hard13.6%
τ²-Bench (Telecom)47.1%
Design Arena (Elo)
Website1,1891,051
UI components1,1801,016
Data viz1,1811,117
SVG1,176
3D1,182878
Game dev1,1951,103
Code categories1,1861,041
ASCII art1,185
Specs
Context window200K1.05M
Max output tokens32K33K
Input modalitiesimage, text, fileimage, text, file
TokenizerClaudeGPT