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 Haiku 4.5GPT-4.1 Mini
Claude Haiku 4.5

Anthropic

GPT-4.1 Mini

OpenAI

Pricing
Input $/1M$1.00$0.40
Output $/1M$5.00$1.60
Blended $/1M (3:1)$2.00$0.70
Cache read $/1M$0.10$0.10
Cache write $/1M$1.25
Quality indices
Intelligence Index29.614.8
Coding Index43.920.2
Math Index46.3
Agentic Index16.41.7
Speed & latency
Output speed (tok/s)83
Time to first token0.55s
Time to first answer token0.55s
Benchmarks (Artificial Analysis)
MMLU-Pro78.1%
GPQA Diamond66.4%
Humanity's Last Exam4.6%
LiveCodeBench48.3%
SciCode40.4%
MATH-50092.5%
AIME43.0%
AIME 202546.3%
IFBench38.3%
AA-LCR (Long Context Reasoning)42.3%
Terminal-Bench Hard7.6%
Terminal-Bench 2.110.1%
τ²-Bench (Telecom)52.9%
τ³-Bench Banking5.2%
Design Arena (Elo)
Website1,1491,025
UI components1,1421,004
Data viz1,1601,070
SVG1,072
3D1,131895
Game dev1,1551,129
Code categories1,1481,027
ASCII art1,182
Specs
Context window200K1.05M
Max output tokens64K33K
Input modalitiestext, image, fileimage, text, file
TokenizerClaudeGPT