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.

GPT-5.6 SolMiMo-V2.5-Pro
GPT-5.6 Sol

OpenAI

MiMo-V2.5-Pro

Xiaomi

Pricing
Input $/1M$5.00$0.435
Output $/1M$30.00$0.87
Blended $/1M (3:1)$11.25$0.544
Cache read $/1M$0.50$0.0036
Cache write $/1M$6.25
Quality indices
Intelligence Index58.942.2
Coding Index77.460.2
Math Index
Agentic Index5429.1
Speed & latency
Output speed (tok/s)6758
Time to first token120.38s2.68s
Time to first answer token120.38s36.90s
Benchmarks (Artificial Analysis)
GPQA Diamond94.1%86.6%
Humanity's Last Exam47.2%33.8%
SciCode56.1%50.2%
IFBench72.7%79.9%
AA-LCR (Long Context Reasoning)73.7%73.3%
Terminal-Bench Hard65.9%43.2%
Terminal-Bench 2.188.0%65.2%
τ²-Bench (Telecom)85.1%94.2%
τ³-Bench Banking33.0%8.7%
Design Arena (Elo)
Website1,296
UI components1,281
Data viz1,281
SVG1,221
3D1,304
Game dev1,331
Code categories1,302
ASCII art1,187
Specs
Context window1.05M1.05M
Max output tokens128K131K
Input modalitiesfile, image, texttext
TokenizerGPTOther