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-5.3-Codex
Claude Haiku 4.5

Anthropic

GPT-5.3-Codex

OpenAI

Pricing
Input $/1M$1.00$1.75
Output $/1M$5.00$14.00
Blended $/1M (3:1)$2.00$4.81
Cache read $/1M$0.10$0.175
Cache write $/1M$1.25
Quality indices
Intelligence Index29.644.3
Coding Index43.9
Math Index
Agentic Index16.4
Speed & latency
Output speed (tok/s)126
Time to first token51.38s
Time to first answer token51.38s
Benchmarks (Artificial Analysis)
GPQA Diamond91.5%
Humanity's Last Exam39.9%
SciCode53.2%
IFBench75.4%
AA-LCR (Long Context Reasoning)74.0%
Terminal-Bench Hard53.0%
τ²-Bench (Telecom)86.0%
Design Arena (Elo)
Website1,1491,190
Web apps1,103
Full-stack1,041
Mobile apps1,127
Android native1,084
UI components1,1421,179
Data viz1,1601,199
SVG1,0721,176
3D1,1311,067
Game dev1,1551,219
Godot game dev1,124
Code categories1,1481,178
ASCII art1,1821,193
Specs
Context window200K400K
Max output tokens64K128K
Input modalitiestext, image, filetext, image, file
TokenizerClaudeGPT