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-oss-20b
Gemini 3.6 Flash

Google

gpt-oss-20b

OpenAI

Pricing
Input $/1M$0.75$0.03
Output $/1M$3.75$0.13
Blended $/1M (3:1)$1.50$0.055
Cache read $/1M$0.075$0.03
Cache write $/1M$0.042
Quality indices
Intelligence Index349
Coding Index69.220.7
Math Index89.3
Agentic Index291.2
Speed & latency
Output speed (tok/s)265189
Time to first token21.43s0.47s
Time to first answer token21.43s11.04s
Benchmarks (Artificial Analysis)
MMLU-Pro74.8%
GPQA Diamond92.8%68.8%
Humanity's Last Exam40.8%11.0%
LiveCodeBench77.7%
SciCode53.4%38.9%
AIME 202589.3%
IFBench65.1%
AA-LCR (Long Context Reasoning)80.0%34.7%
Terminal-Bench Hard10.6%
Terminal-Bench 2.177.5%13.9%
τ²-Bench (Telecom)60.2%
τ³-Bench Banking29.9%7.0%
Design Arena (Elo)
Website1,312865
Web apps1,211
Full-stack1,188
Mobile apps1,228
Android native1,213
UI components1,318
Data viz1,311946
3D1,302
Game dev1,286
Agentic game dev1,180
HTML slides1,152
Python→PPTX slides1,146
Code categories1,305
ASCII art1,304
Specs
Context window1.05M131K
Max output tokens66K118K
Input modalitiestext, image, video, file, audiotext
TokenizerGeminiGPT