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.

MiniMax M3GPT-5.6 Sol
MiniMax M3

MiniMax

GPT-5.6 Sol

OpenAI

Pricing
Input $/1M$0.30$5.00
Output $/1M$1.20$30.00
Blended $/1M (3:1)$0.525$11.25
Cache read $/1M$0.06$0.50
Cache write $/1M$6.25
Quality indices
Intelligence Index44.458.9
Coding Index58.677.4
Math Index
Agentic Index35.454
Speed & latency
Output speed (tok/s)9767
Time to first token1.29s120.38s
Time to first answer token21.90s120.38s
Benchmarks (Artificial Analysis)
GPQA Diamond92.9%94.1%
Humanity's Last Exam37.1%47.2%
SciCode45.4%56.1%
IFBench82.9%72.7%
AA-LCR (Long Context Reasoning)74.0%73.7%
Terminal-Bench Hard42.4%65.9%
Terminal-Bench 2.165.2%88.0%
τ²-Bench (Telecom)88.9%85.1%
τ³-Bench Banking13.0%33.0%
Design Arena (Elo)
Website1,287
Web apps1,263
Full-stack1,250
Mobile apps1,256
Android native1,099
UI components1,280
Data viz1,277
SVG1,227
3D1,280
Game dev1,280
Code categories1,287
ASCII art1,198
Specs
Context window1.05M1.05M
Max output tokens512K128K
Input modalitiestext, image, videofile, image, text
TokenizerOtherGPT