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
Gemini 3.6 Flash

Google

GPT-5

OpenAI

Pricing
Input $/1M$1.50$1.25
Output $/1M$7.50$10.00
Blended $/1M (3:1)$3.00$3.44
Cache read $/1M$0.15$0.125
Cache write $/1M$0.083
Quality indices
Intelligence Index50.134.7
Coding Index69.237.8
Math Index94.3
Agentic Index38.725.7
Speed & latency
Output speed (tok/s)311109
Time to first token12.80s96.67s
Time to first answer token12.80s96.67s
Benchmarks (Artificial Analysis)
MMLU-Pro87.1%
GPQA Diamond92.8%85.4%
Humanity's Last Exam38.3%26.5%
LiveCodeBench84.6%
SciCode52.7%42.9%
MATH-50099.4%
AIME95.7%
AIME 202594.3%
IFBench73.1%
AA-LCR (Long Context Reasoning)69.7%75.6%
Terminal-Bench Hard32.6%
Terminal-Bench 2.177.5%35.2%
τ²-Bench (Telecom)84.8%
τ³-Bench Banking24.5%19.6%
Design Arena (Elo)
Website1,3321,212
UI components1,224
Data viz1,275
SVG1,235
3D1,115
Game dev1,245
Code categories1,202
ASCII art1,183
Specs
Context window1.05M400K
Max output tokens66K128K
Input modalitiestext, image, video, file, audiotext, image, file
TokenizerGeminiGPT