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.

Gemini 3.6 FlashGPT-5.2 (batch)
Gemini 3.6 Flash

Google

GPT-5.2 (batch)

OpenAI

Pricing
Input $/1M$0.75$0.875
Output $/1M$3.75$7.00
Blended $/1M (3:1)$1.50$2.41
Cache read $/1M$0.075$0.087
Cache write $/1M$0.042
Quality indices
Intelligence Index3430.4
Coding Index69.2
Math Index99
Agentic Index29
Speed & latency
Output speed (tok/s)2650
Time to first token21.43s0s
Time to first answer token21.43s0s
Benchmarks (Artificial Analysis)
MMLU-Pro87.4%
GPQA Diamond92.8%90.3%
Humanity's Last Exam40.8%37.7%
LiveCodeBench88.9%
SciCode53.4%
AIME 202599.0%
IFBench75.4%
AA-LCR (Long Context Reasoning)80.0%82.7%
Terminal-Bench Hard47.0%
Terminal-Bench 2.177.5%
τ²-Bench (Telecom)84.8%
τ³-Bench Banking29.9%
Design Arena (Elo)
Website1,3121,207
Web apps1,2111,101
Full-stack1,1881,052
Mobile apps1,2281,123
Android native1,2131,073
UI components1,3181,206
Data viz1,3111,216
SVG1,162
3D1,3021,108
Game dev1,2861,219
Agentic game dev1,180
Godot game dev1,142
HTML slides1,152
Python→PPTX slides1,146
Code categories1,3051,185
ASCII art1,3041,171
Specs
Context window1.05M400K
Max output tokens66K128K
Input modalitiestext, image, video, file, audiofile, image, text
TokenizerGeminiGPT