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-5.2 (batch)
Claude Sonnet 5

Anthropic

GPT-5.2 (batch)

OpenAI

Pricing
Input $/1M$2.00$0.875
Output $/1M$10.00$7.00
Blended $/1M (3:1)$4.00$2.41
Cache read $/1M$0.20$0.087
Cache write $/1M$2.50
Quality indices
Intelligence Index38.230.4
Coding Index71.5
Math Index99
Agentic Index43.6
Speed & latency
Output speed (tok/s)820
Time to first token118.14s0s
Time to first answer token118.14s0s
Benchmarks (Artificial Analysis)
MMLU-Pro87.4%
GPQA Diamond91.1%90.3%
Humanity's Last Exam41.3%37.7%
LiveCodeBench88.9%
SciCode54.3%
AIME 202599.0%
IFBench75.4%
AA-LCR (Long Context Reasoning)82.0%82.7%
Terminal-Bench Hard47.0%
Terminal-Bench 2.180.5%
τ²-Bench (Telecom)84.8%
τ³-Bench Banking37.3%
Design Arena (Elo)
Website1,2881,207
Web apps1,2601,101
Full-stack1,2601,052
Mobile apps1,2491,123
Android native1,2401,073
UI components1,2991,206
Data viz1,2601,216
SVG1,2121,162
3D1,2891,108
Game dev1,3131,219
Agentic game dev1,227
Godot game dev1,2381,142
HTML slides1,220
Python→PPTX slides1,216
Code categories1,2931,185
ASCII art1,2261,171
Specs
Context window1M400K
Max output tokens128K128K
Input modalitiestext, image, filefile, image, text
TokenizerClaudeGPT