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 Flasho3 Mini High
Gemini 3.6 Flash

Google

o3 Mini High

OpenAI

Pricing
Input $/1M$1.50$1.10
Output $/1M$7.50$4.40
Blended $/1M (3:1)$3.00$1.93
Cache read $/1M$0.15$0.55
Cache write $/1M$0.083
Quality indices
Intelligence Index50.115.6
Coding Index69.216.3
Math Index
Agentic Index38.71.7
Speed & latency
Output speed (tok/s)311239
Time to first token12.80s14.66s
Time to first answer token12.80s14.66s
Benchmarks (Artificial Analysis)
MMLU-Pro80.2%
GPQA Diamond92.8%77.3%
Humanity's Last Exam38.3%12.3%
LiveCodeBench73.4%
SciCode52.7%39.8%
MATH-50098.5%
AIME86.0%
IFBench67.1%
AA-LCR (Long Context Reasoning)69.7%39.3%
Terminal-Bench Hard6.1%
Terminal-Bench 2.177.5%4.5%
τ²-Bench (Telecom)31.3%
τ³-Bench Banking24.5%5.2%
Design Arena (Elo)
Website1,332
Specs
Context window1.05M200K
Max output tokens66K100K
Input modalitiestext, image, video, file, audiotext, file
TokenizerGeminiGPT